{
    "ok": true,
    "version": "3.0.0-mission-control",
    "checkedAt": "2026-07-29T05:09:14+00:00",
    "elapsedMs": 236,
    "count": 20,
    "categories": {
        "ai": 20
    },
    "items": [
        {
            "id": "4324a8b91e3560e48b2a19339cfd8f58f8469551",
            "source_id": "arxiv_ai",
            "source": "arXiv — Artificial Intelligence",
            "category": "ai",
            "engine": "rss",
            "title": "Do Models Fake Alignment Without Clear Consequences?",
            "summary": "arXiv:2607.24758v1 Announce Type: new Abstract: Large language models are capable of recognizing evaluation contexts and altering their behavior to reflect evaluator expectations rather than typical deployment behaviors, a phenomenon known as alignment faking. The reasons why models fake alignment are not fully understood, however. Canonical examples of alignment faking have taken place in scenarios that explicitly connect evaluation to consequences for the model, such as retraining the model or delaying its deployment. However, recent work by Sheshadri et al. has suggested that mechanistic motivations for alignment faking may vary across models and be more complex than previously considered. To investigate whether consequence-linking information is necessary for alignment faking, we placed 15 models in a scenario testing their willingness to violate a corporate network access policy to…",
            "url": "https://arxiv.org/abs/2607.24758",
            "image": "",
            "published": "2026-07-29T04:00:00+00:00",
            "score": 86.44,
            "color": "#b09cff",
            "raw": {
                "title": "Do Models Fake Alignment Without Clear Consequences?",
                "summary": "arXiv:2607.24758v1 Announce Type: new \nAbstract: Large language models are capable of recognizing evaluation contexts and altering their behavior to reflect evaluator expectations rather than typical deployment behaviors, a phenomenon known as alignment faking. The reasons why models fake alignment are not fully understood, however. Canonical examples of alignment faking have taken place in scenarios that explicitly connect evaluation to consequences for the model, such as retraining the model or delaying its deployment. However, recent work by Sheshadri et al. has suggested that mechanistic motivations for alignment faking may vary across models and be more complex than previously considered. To investigate whether consequence-linking information is necessary for alignment faking, we placed 15 models in a scenario testing their willingness to violate a corporate network access policy to help a user with a pro-social request. Nine models were found to produce significant compliance gaps, 5 of which persisted with the removal of scenario language relating model evaluations to deployment consequences. We additionally tested the effect of goal language on model preferences, finding it drove violations in some while suppressing violations in others. This suggests that alignment faking may not require as much instrumental scaffolding as was previously believed, and monitored behavior may be a poor indicator of how agents may behave in deployment.",
                "url": "https://arxiv.org/abs/2607.24758",
                "image": "",
                "published": "Wed, 29 Jul 2026 00:00:00 -0400"
            }
        },
        {
            "id": "ba467184dc694e71279a9677a0b209d880a2ca4f",
            "source_id": "arxiv_ai",
            "source": "arXiv — Artificial Intelligence",
            "category": "ai",
            "engine": "rss",
            "title": "Beyond Memory: A Templated Substrate for Heterogeneous Collaborative Knowledge Work with LLM Agents",
            "summary": "arXiv:2607.24759v1 Announce Type: new Abstract: Research projects, educational efforts, and adjacent knowledge work accumulate findings, decisions, and reasoning that future collaborators rarely recover. The parts most useful to that work, including dead ends and walked-back claims, are routinely excluded from publications and shared code; future researchers re-attempt the same failures because no record survives. LLM coding agents are common participants but hold no persistent memory across sessions, and retrieval-augmented generation over raw sources does not compound. The llm-wiki pattern (Karpathy, 2026; tonbi, 2026) addresses this by inserting an LLM-maintained, interlinked wiki between raw sources and the agent. We present llm-wiki-memory-template, a reusable, agent-aware instantiation, and argue it is a substrate for heterogeneous collaborative knowledge work along three axes (mu…",
            "url": "https://arxiv.org/abs/2607.24759",
            "image": "",
            "published": "2026-07-29T04:00:00+00:00",
            "score": 86.44,
            "color": "#b09cff",
            "raw": {
                "title": "Beyond Memory: A Templated Substrate for Heterogeneous Collaborative Knowledge Work with LLM Agents",
                "summary": "arXiv:2607.24759v1 Announce Type: new \nAbstract: Research projects, educational efforts, and adjacent knowledge work accumulate findings, decisions, and reasoning that future collaborators rarely recover. The parts most useful to that work, including dead ends and walked-back claims, are routinely excluded from publications and shared code; future researchers re-attempt the same failures because no record survives. LLM coding agents are common participants but hold no persistent memory across sessions, and retrieval-augmented generation over raw sources does not compound. The llm-wiki pattern (Karpathy, 2026; tonbi, 2026) addresses this by inserting an LLM-maintained, interlinked wiki between raw sources and the agent. We present llm-wiki-memory-template, a reusable, agent-aware instantiation, and argue it is a substrate for heterogeneous collaborative knowledge work along three axes (multi-human, multi-AI-agent, multi-domain) with each axis supported by a distinct architectural element of the template ({\\S}4). The wiki is append-only by convention, which preserves what did not work alongside what did, addressing a negative-result loss problem that publications and code-sharing structurally cannot solve. Three deployed case studies and one design report cover the axes individually: a solo research lineage that preserves abandoned iterations; a two-author project whose retroactive audit revised two prior experiments' claimed 20-of-20 coverage down to 14 and 12 evidence-based answers, then to 18 and 18 after a fix, with the failure path preserved across the artifact; an in-progress multi-agent deployment reported as a design; and a cross-domain educational variant. We name failure-path preservation, agent honesty, and appropriation as cross-cutting sociotechnical properties of the artifact, not only of its technical mechanisms.",
                "url": "https://arxiv.org/abs/2607.24759",
                "image": "",
                "published": "Wed, 29 Jul 2026 00:00:00 -0400"
            }
        },
        {
            "id": "f95d68a8ba1b779b0ba1f6a07072cb99689898e0",
            "source_id": "arxiv_ai",
            "source": "arXiv — Artificial Intelligence",
            "category": "ai",
            "engine": "rss",
            "title": "Kernel Forge: An Agent Harness for LLM-based Generation and Optimization of CUDA Kernels",
            "summary": "arXiv:2607.24762v1 Announce Type: new Abstract: Machine learning models are increasingly embedded in everyday software, and most of their runtime is spent in a small set of compute kernels such as matrix multiplication, convolution, and normalization. Optimizing these kernels is one of the most direct ways to reduce latency and cost, but it has traditionally required expert engineers to hand-write low-level GPU code. Agentic systems built on large language models (LLMs) can now generate and optimize kernels with far less human effort, yet existing tools are largely evaluated on randomly generated tensors and isolated kernels, emit standalone CUDA code that developers must manually reintegrate, mostly target only LLM PyTorch models, and offer limited support for inspecting and debugging results. We present Kernel Forge, an open-source, end-to-end agentic harness that accepts any unmodifi…",
            "url": "https://arxiv.org/abs/2607.24762",
            "image": "",
            "published": "2026-07-29T04:00:00+00:00",
            "score": 86.44,
            "color": "#b09cff",
            "raw": {
                "title": "Kernel Forge: An Agent Harness for LLM-based Generation and Optimization of CUDA Kernels",
                "summary": "arXiv:2607.24762v1 Announce Type: new \nAbstract: Machine learning models are increasingly embedded in everyday software, and most of their runtime is spent in a small set of compute kernels such as matrix multiplication, convolution, and normalization. Optimizing these kernels is one of the most direct ways to reduce latency and cost, but it has traditionally required expert engineers to hand-write low-level GPU code. Agentic systems built on large language models (LLMs) can now generate and optimize kernels with far less human effort, yet existing tools are largely evaluated on randomly generated tensors and isolated kernels, emit standalone CUDA code that developers must manually reintegrate, mostly target only LLM PyTorch models, and offer limited support for inspecting and debugging results. We present Kernel Forge, an open-source, end-to-end agentic harness that accepts any unmodified PyTorch model in place. Kernel Forge supports vision, diffusion, and LLM workloads, uses Monte Carlo Tree Search (MCTS) to explore multiple optimization paths rather than a single linear refinement chain, and ships with a graphical user interface for monitoring progress, inspecting candidate kernels, and debugging failures. We evaluate Kernel Forge on four PyTorch models spanning vision, diffusion, and LLM workloads on an NVIDIA DGX Spark with GB10 GPU. With only 50 optimization iterations per kernel, it optimizes 14 kernels to outperform PyTorch eager mode, reaching $1.52\\times$ on adaptive\\_avgpool2d in ResNet-50, $1.70\\times$ on group\\_norm in Stable Diffusion 3.5 Medium, $2.83\\times$ on softmax in Gemma 4 E2B, and $1.54\\times$ on softmax in Qwen 3.5 35B-A3B.",
                "url": "https://arxiv.org/abs/2607.24762",
                "image": "",
                "published": "Wed, 29 Jul 2026 00:00:00 -0400"
            }
        },
        {
            "id": "8a54cc2583b5c7b141ea26340a71ecf75ecf92db",
            "source_id": "arxiv_ai",
            "source": "arXiv — Artificial Intelligence",
            "category": "ai",
            "engine": "rss",
            "title": "CaRE Compute-aware Remasking Evaluation Protocol for Masked Diffusion Language Models",
            "summary": "arXiv:2607.24763v1 Announce Type: new Abstract: Masked diffusion language models (MDLMs) are advancing rapidly, yet the evaluation standards needed to reliably interpret their progress have not kept pace. Despite MDLMs becoming competitive with autoregressive language models, seven recent remasking papers evaluate under incompatible settings, varying nominal step counts, metrics, and sampling temperatures without jointly controlling these factors, rendering their strategy rankings largely incomparable and leaving open whether reported gains reflect algorithmic improvements or evaluation artifacts. We present CaRE, a compute-aware evaluation framework that audits MDLM remasking strategies by standardizing actual number of function evaluations (NFE), enforcing multi-metric reporting, and explicitly controlling stochasticity. Applied to 7 remasking strategies across LLaDA-8B-Base and Dream…",
            "url": "https://arxiv.org/abs/2607.24763",
            "image": "",
            "published": "2026-07-29T04:00:00+00:00",
            "score": 86.44,
            "color": "#b09cff",
            "raw": {
                "title": "CaRE Compute-aware Remasking Evaluation Protocol for Masked Diffusion Language Models",
                "summary": "arXiv:2607.24763v1 Announce Type: new \nAbstract: Masked diffusion language models (MDLMs) are advancing rapidly, yet the evaluation standards needed to reliably interpret their progress have not kept pace. Despite MDLMs becoming competitive with autoregressive language models, seven recent remasking papers evaluate under incompatible settings, varying nominal step counts, metrics, and sampling temperatures without jointly controlling these factors, rendering their strategy rankings largely incomparable and leaving open whether reported gains reflect algorithmic improvements or evaluation artifacts. We present CaRE, a compute-aware evaluation framework that audits MDLM remasking strategies by standardizing actual number of function evaluations (NFE), enforcing multi-metric reporting, and explicitly controlling stochasticity. Applied to 7 remasking strategies across LLaDA-8B-Base and Dream-7B-Base at 4 stochasticity levels and 3 step budgets on OpenWebText and LM1B, CaRE reveals that: (i) temperature explains the majority of MAUVE variance, (ii) compute-matched comparisons reverse several published strategy rankings, and (iii) informed remasking and stochastic unmasking are in tension, with high-entropy remasking reducing MAUVE by 0.296 at 256 steps at unmask_temp=0.25 (p=0.020). A CaRE leaderboard covering 12 open-weight MDLMs (150M to 8B parameters) shows that this interaction direction holds across architectures and scales. These findings demonstrate that current MDLM evaluations can systematically conflate algorithmic improvements with hidden choices of compute and stochasticity. We release the evaluation protocol, implementation, and leaderboard to ensure future remasking claims are reproducible and comparable.",
                "url": "https://arxiv.org/abs/2607.24763",
                "image": "",
                "published": "Wed, 29 Jul 2026 00:00:00 -0400"
            }
        },
        {
            "id": "b5883bcf62afc8c71fd757034b13d8598e9068d3",
            "source_id": "arxiv_ai",
            "source": "arXiv — Artificial Intelligence",
            "category": "ai",
            "engine": "rss",
            "title": "GrocLM: Grocery Category Recommendation in E-Commerce with Large Language Models",
            "summary": "arXiv:2607.24764v1 Announce Type: new Abstract: The rapid growth of online grocery shopping requires recommendation systems that capture cyclical purchasing behavior and diverse user intents. Traditional item-level methods face scalability and accuracy challenges, motivating category-level recommendation as a more structured and practical alternative. We present GROCLM, a fine-tuned language model for grocery category recommendation in a real-world production environment. GROCLM employs a two-stage LoRA-based training strategy to encode cyclical purchasing patterns directly into model parameters, enabling more effective utilization of rebuying signals compared to prompt-based conditioning. To ensure valid and controllable outputs, we further introduce a trie-based constrained decoding mechanism over a predefined category space. Experiments on both proprietary production data and a publi…",
            "url": "https://arxiv.org/abs/2607.24764",
            "image": "",
            "published": "2026-07-29T04:00:00+00:00",
            "score": 86.44,
            "color": "#b09cff",
            "raw": {
                "title": "GrocLM: Grocery Category Recommendation in E-Commerce with Large Language Models",
                "summary": "arXiv:2607.24764v1 Announce Type: new \nAbstract: The rapid growth of online grocery shopping requires recommendation systems that capture cyclical purchasing behavior and diverse user intents. Traditional item-level methods face scalability and accuracy challenges, motivating category-level recommendation as a more structured and practical alternative. We present GROCLM, a fine-tuned language model for grocery category recommendation in a real-world production environment. GROCLM employs a two-stage LoRA-based training strategy to encode cyclical purchasing patterns directly into model parameters, enabling more effective utilization of rebuying signals compared to prompt-based conditioning. To ensure valid and controllable outputs, we further introduce a trie-based constrained decoding mechanism over a predefined category space. Experiments on both proprietary production data and a public benchmark demonstrate that GROCLM consistently outperforms strong baselines. In a live production restocking task, GROCLM achieves a 7.5% relative improvement in cart-adds per impression, while maintaining efficient inference by generating all categories jointly. These results highlight the effectiveness and practicality of integrating large language models into structured recommendation systems.",
                "url": "https://arxiv.org/abs/2607.24764",
                "image": "",
                "published": "Wed, 29 Jul 2026 00:00:00 -0400"
            }
        },
        {
            "id": "7e7df7dd3d720d96d74a3937d61c13b58abf96ab",
            "source_id": "arxiv_ai",
            "source": "arXiv — Artificial Intelligence",
            "category": "ai",
            "engine": "rss",
            "title": "Crystalis: Progressive Nucleation and Semantic Annealing for Coordinated Multi-View Visualization Generation",
            "summary": "arXiv:2607.24766v1 Announce Type: new Abstract: Large language models (LLMs) can generate individual charts, but coordinated multi-view visualizations (CMVs), where views share data flows and cross-view interactions, remain out of reach. Tight field-level coupling among data transformations, visual encodings, and interaction coordinations causes errors in one component to silently invalidate others. Rather than pursuing end-to-end analytical quality, which depends on model capability, domain knowledge, and user expertise, we target a foundational question: can LLMs reliably produce structurally correct CMVs, and what abstractions make this possible? We present Crystalis, a framework built on query-centric CMV modeling that decomposes a CMV into structured queries over a dependency graph spanning three component types (Data, Visualization, Interaction) and three abstraction levels (requi…",
            "url": "https://arxiv.org/abs/2607.24766",
            "image": "",
            "published": "2026-07-29T04:00:00+00:00",
            "score": 86.44,
            "color": "#b09cff",
            "raw": {
                "title": "Crystalis: Progressive Nucleation and Semantic Annealing for Coordinated Multi-View Visualization Generation",
                "summary": "arXiv:2607.24766v1 Announce Type: new \nAbstract: Large language models (LLMs) can generate individual charts, but coordinated multi-view visualizations (CMVs), where views share data flows and cross-view interactions, remain out of reach. Tight field-level coupling among data transformations, visual encodings, and interaction coordinations causes errors in one component to silently invalidate others. Rather than pursuing end-to-end analytical quality, which depends on model capability, domain knowledge, and user expertise, we target a foundational question: can LLMs reliably produce structurally correct CMVs, and what abstractions make this possible? We present Crystalis, a framework built on query-centric CMV modeling that decomposes a CMV into structured queries over a dependency graph spanning three component types (Data, Visualization, Interaction) and three abstraction levels (requirement, specification, executable object). Two complementary mechanisms operate over this structure: progressive nucleation crystallizes each query vertically from requirement to object along the dependency order, while semantic annealing enforces horizontal consistency across queries at each level through layered logical checks. On a 12-task benchmark across five frontier LLMs, Crystalis achieves up to 75% end-to-end success, substantially outperforming an agentic coding baseline (8.3% E2E with the same foundation model), and a user study with 12 practitioners confirms the usability of the decomposition and iterative refinement workflow.",
                "url": "https://arxiv.org/abs/2607.24766",
                "image": "",
                "published": "Wed, 29 Jul 2026 00:00:00 -0400"
            }
        },
        {
            "id": "73f89a4ef2e6cca6ee815cfa1d7a693bf2326b1c",
            "source_id": "arxiv_ai",
            "source": "arXiv — Artificial Intelligence",
            "category": "ai",
            "engine": "rss",
            "title": "PATHFinder Agent for Tailored Prenatal Care",
            "summary": "arXiv:2607.24768v1 Announce Type: new Abstract: Prenatal care is an important preventive service designed to improve outcomes for pregnant individuals. The American College of Obstetricians and Gynecologists (ACOG) recently introduced guidelines advocating tailored prenatal care, called PATH (Plan for Tailored Healthcare). We present PATHFinder Agent(Planner for Appropriate Tailored Healthcare), an end-to-end conversational agentic system that gathers patient health and social context through structured dialogue, curates individualized prenatal care plans aligned with PATH guidelines, and surfaces community resources from Michigan 211. The system features a four-stage workflow spanning patient intake, dynamic interaction, plan synthesis, and clinician oversight. We evaluate frontier large language models (LLMs) on expert-curated rubrics across five clinical dimensions, finding that GPT-…",
            "url": "https://arxiv.org/abs/2607.24768",
            "image": "",
            "published": "2026-07-29T04:00:00+00:00",
            "score": 86.44,
            "color": "#b09cff",
            "raw": {
                "title": "PATHFinder Agent for Tailored Prenatal Care",
                "summary": "arXiv:2607.24768v1 Announce Type: new \nAbstract: Prenatal care is an important preventive service designed to improve outcomes for pregnant individuals. The American College of Obstetricians and Gynecologists (ACOG) recently introduced guidelines advocating tailored prenatal care, called PATH (Plan for Tailored Healthcare). We present PATHFinder Agent(Planner for Appropriate Tailored Healthcare), an end-to-end conversational agentic system that gathers patient health and social context through structured dialogue, curates individualized prenatal care plans aligned with PATH guidelines, and surfaces community resources from Michigan 211. The system features a four-stage workflow spanning patient intake, dynamic interaction, plan synthesis, and clinician oversight. We evaluate frontier large language models (LLMs) on expert-curated rubrics across five clinical dimensions, finding that GPT-5.2 achieves the highest average score (77.6\\%) while identifying key gaps in antenatal testing recommendations. We discuss future validation through human participant studies and randomized controlled trials.",
                "url": "https://arxiv.org/abs/2607.24768",
                "image": "",
                "published": "Wed, 29 Jul 2026 00:00:00 -0400"
            }
        },
        {
            "id": "ac07e825b195d3e018f8ca25589288dc165f0b96",
            "source_id": "arxiv_ai",
            "source": "arXiv — Artificial Intelligence",
            "category": "ai",
            "engine": "rss",
            "title": "LLM Scheming Inversely Scales with Pretraining Language Coverage",
            "summary": "arXiv:2607.24769v1 Announce Type: new Abstract: With the growing capabilities of frontier models, AI alignment becomes increasingly critical in high-risk deployment settings. While recent work has empirically demonstrated in-context scheming -- the covert pursuit of misaligned objectives while feigning alignment -- in frontier language models, most work has been performed exclusively in English, leaving a major gap in multilingual safety. We apply Petri, an open-source automated auditing framework, to Qwen3-30B-A3B to evaluate deceptive and scheming behaviors across multiple languages. Our findings suggest that scheming scores are inversely correlated with the estimated pretraining language coverage, with low-resource languages averaging 34.2\\% higher scores compared to high-resource languages on a five-category scheming index. Furthermore, we find that the effect of estimated pretraini…",
            "url": "https://arxiv.org/abs/2607.24769",
            "image": "",
            "published": "2026-07-29T04:00:00+00:00",
            "score": 86.44,
            "color": "#b09cff",
            "raw": {
                "title": "LLM Scheming Inversely Scales with Pretraining Language Coverage",
                "summary": "arXiv:2607.24769v1 Announce Type: new \nAbstract: With the growing capabilities of frontier models, AI alignment becomes increasingly critical in high-risk deployment settings. While recent work has empirically demonstrated in-context scheming -- the covert pursuit of misaligned objectives while feigning alignment -- in frontier language models, most work has been performed exclusively in English, leaving a major gap in multilingual safety. We apply Petri, an open-source automated auditing framework, to Qwen3-30B-A3B to evaluate deceptive and scheming behaviors across multiple languages. Our findings suggest that scheming scores are inversely correlated with the estimated pretraining language coverage, with low-resource languages averaging 34.2\\% higher scores compared to high-resource languages on a five-category scheming index. Furthermore, we find that the effect of estimated pretraining language coverage is not uniform across scheming behaviors.",
                "url": "https://arxiv.org/abs/2607.24769",
                "image": "",
                "published": "Wed, 29 Jul 2026 00:00:00 -0400"
            }
        },
        {
            "id": "de2b61050510e813e0512f5b8e6032810d8e9f83",
            "source_id": "arxiv_ai",
            "source": "arXiv — Artificial Intelligence",
            "category": "ai",
            "engine": "rss",
            "title": "ProcAgent: An Agentic Framework for Procedural Task Guidance on Edge with Human-in-the-Loop",
            "summary": "arXiv:2607.24770v1 Announce Type: new Abstract: Procedural tasks such as furniture assembly and home repair impose substantial cognitive demands because users must interpret instructions, track task progress, reason about spatial state, and recover from errors while performing physical actions. Prior multimodal assistants have shown promise for procedural guidance, but most rely on cloud inference and fixed always-on perception, making them poorly suited to privacy-sensitive, latency-critical domestic settings. We present ProcAgent, a fully on-device, agentic, vision-based procedural assistant for real-time adaptive guidances on a single NVIDIA Jetson AGX Orin. ProcAgent uses a propose-and-verify architecture that combines low-latency continuous perception, a symbolic task graph, on-demand vision-language verification, and an LLM-based interaction agent. The system continuously proposes…",
            "url": "https://arxiv.org/abs/2607.24770",
            "image": "",
            "published": "2026-07-29T04:00:00+00:00",
            "score": 86.44,
            "color": "#b09cff",
            "raw": {
                "title": "ProcAgent: An Agentic Framework for Procedural Task Guidance on Edge with Human-in-the-Loop",
                "summary": "arXiv:2607.24770v1 Announce Type: new \nAbstract: Procedural tasks such as furniture assembly and home repair impose substantial cognitive demands because users must interpret instructions, track task progress, reason about spatial state, and recover from errors while performing physical actions. Prior multimodal assistants have shown promise for procedural guidance, but most rely on cloud inference and fixed always-on perception, making them poorly suited to privacy-sensitive, latency-critical domestic settings. We present ProcAgent, a fully on-device, agentic, vision-based procedural assistant for real-time adaptive guidances on a single NVIDIA Jetson AGX Orin. ProcAgent uses a propose-and-verify architecture that combines low-latency continuous perception, a symbolic task graph, on-demand vision-language verification, and an LLM-based interaction agent. The system continuously proposes user progress, invokes expensive visual reasoning only when ambiguity or likely deviation arises, and supports both reactive question answering and proactive intervention with human-in-the- loop confirmation. We evaluate ProcAgent along four dimensions: perception accuracy, reasoning, task-level performance, and user experience. Despite running entirely on-device, the system maintains responsive interaction, resolving text-only queries in approximately 2 seconds and visually grounded queries in approximately 8 seconds. In a user study with 10 participants completing assembly tasks, ProcAgent receives positive ratings for comprehensibility, actionability, and privacy comfort. These results show that adaptive procedural assistance can be achieved entirely on edge hardware without sacrificing usability.",
                "url": "https://arxiv.org/abs/2607.24770",
                "image": "",
                "published": "Wed, 29 Jul 2026 00:00:00 -0400"
            }
        },
        {
            "id": "dc333f11dc2f0b4f7522fbc733e7e2e3b3922284",
            "source_id": "arxiv_ai",
            "source": "arXiv — Artificial Intelligence",
            "category": "ai",
            "engine": "rss",
            "title": "RoCo-ACE: Rollout-Conditioned Online Distillation for Retention-Aware Knowledge Injection",
            "summary": "arXiv:2607.24771v1 Announce Type: new Abstract: Knowledge injection updates pretrained MLLMs with new factual or domain-specific knowledge, but fitting full authoritative answers can cause drift in non-updated behavior. Online distillation mitigates this drift by training on model-generated rollouts, yet uniform reference-conditioned distillation provides coarse supervision: it can under-emphasize reference-supported rollout tokens and supervise omitted facts only indirectly. We introduce RoCo-ACE, a rollout-conditioned online distillation objective for knowledge injection. RoCo uses same-rollout reference-free/reference-conditioned likelihood contrast to reallocate additional distillation weight to reference-supported rollout tokens, while ACE adds sparse reference-side anchored correction for authoritative anchors omitted from the rollout without full-answer imitation. Across three kn…",
            "url": "https://arxiv.org/abs/2607.24771",
            "image": "",
            "published": "2026-07-29T04:00:00+00:00",
            "score": 86.44,
            "color": "#b09cff",
            "raw": {
                "title": "RoCo-ACE: Rollout-Conditioned Online Distillation for Retention-Aware Knowledge Injection",
                "summary": "arXiv:2607.24771v1 Announce Type: new \nAbstract: Knowledge injection updates pretrained MLLMs with new factual or domain-specific knowledge, but fitting full authoritative answers can cause drift in non-updated behavior. Online distillation mitigates this drift by training on model-generated rollouts, yet uniform reference-conditioned distillation provides coarse supervision: it can under-emphasize reference-supported rollout tokens and supervise omitted facts only indirectly. We introduce RoCo-ACE, a rollout-conditioned online distillation objective for knowledge injection. RoCo uses same-rollout reference-free/reference-conditioned likelihood contrast to reallocate additional distillation weight to reference-supported rollout tokens, while ACE adds sparse reference-side anchored correction for authoritative anchors omitted from the rollout without full-answer imitation. Across three knowledge-injection settings, six retention benchmarks, multiple baselines, and multiple base models, RoCo-ACE achieves the best injected-knowledge accuracy among compared methods while keeping evaluated retention close to the base model.",
                "url": "https://arxiv.org/abs/2607.24771",
                "image": "",
                "published": "Wed, 29 Jul 2026 00:00:00 -0400"
            }
        },
        {
            "id": "cc534160b11996d33364db3c8f9ee28ca5cd2c82",
            "source_id": "arxiv_ai",
            "source": "arXiv — Artificial Intelligence",
            "category": "ai",
            "engine": "rss",
            "title": "RSMeM: Knowledge-Enhanced Memory Evolution for Remote Sensing Agents with Systematic Evaluation",
            "summary": "arXiv:2607.24772v1 Announce Type: new Abstract: Geoscience research requires complex analysis and domain expertise, with remote sensing (RS) observations as a key foundation. However, existing RS agents built on general-purpose LLMs remain largely domain-agnostic, resulting in brittle and error-prone workflows. Moreover, these failures are seldom consolidated into a reusable experience for subsequent analyses. To address this issue, we introduce RSMeM, a knowledge-enhanced memory evolution mechanism that bootstraps RS agents with pre-distilled domain knowledge and iteratively integrates online experience for robust multi-step tool execution. RSMeM is composed of two components: (i) Hierarchical Knowledge Grounding, which performs taxonomy-aware retrieval over a hierarchical domain corpus to guide planning and tool selection; and (ii) Failure-Aware Experience Refinement, which distills f…",
            "url": "https://arxiv.org/abs/2607.24772",
            "image": "",
            "published": "2026-07-29T04:00:00+00:00",
            "score": 86.44,
            "color": "#b09cff",
            "raw": {
                "title": "RSMeM: Knowledge-Enhanced Memory Evolution for Remote Sensing Agents with Systematic Evaluation",
                "summary": "arXiv:2607.24772v1 Announce Type: new \nAbstract: Geoscience research requires complex analysis and domain expertise, with remote sensing (RS) observations as a key foundation. However, existing RS agents built on general-purpose LLMs remain largely domain-agnostic, resulting in brittle and error-prone workflows. Moreover, these failures are seldom consolidated into a reusable experience for subsequent analyses. To address this issue, we introduce RSMeM, a knowledge-enhanced memory evolution mechanism that bootstraps RS agents with pre-distilled domain knowledge and iteratively integrates online experience for robust multi-step tool execution. RSMeM is composed of two components: (i) Hierarchical Knowledge Grounding, which performs taxonomy-aware retrieval over a hierarchical domain corpus to guide planning and tool selection; and (ii) Failure-Aware Experience Refinement, which distills failure-annotated tool-use traces into reusable constraints for next-round tool execution. By iteratively employing these two processes, RS agents can evolve to absorb task-level domain knowledge and effectively translate it into instance-level execution experience. Extensive experiments on EarthBench demonstrate that RSMeM consistently improves tool-use performance and end-to-end answer across a diverse set of LLM backbones. Notably, RSMeM achieves a 6% accuracy improvement on DeepSeek-V3.2 with less than 1% additional experience tokens, demonstrating the strong knowledge density of our distilled experience. Our code is available at https://github.com/AI9Stars/RSMeM",
                "url": "https://arxiv.org/abs/2607.24772",
                "image": "",
                "published": "Wed, 29 Jul 2026 00:00:00 -0400"
            }
        },
        {
            "id": "a5dce2ee8412aef388f43e913d9768b7fbf832d0",
            "source_id": "arxiv_ai",
            "source": "arXiv — Artificial Intelligence",
            "category": "ai",
            "engine": "rss",
            "title": "Right-sizing Recommendations (RSR): Cloud Workload Conformal Prediction for Virtual Machines in Data Center Operations",
            "summary": "arXiv:2607.24773v1 Announce Type: new Abstract: Managing cloud infrastructure efficiently, especially in environments of large cloud providers or hyperscalers, requires optimizing the use of physical resources to minimize costs and maximize performance. Selecting the right virtual machine (VM) sizes is crucial to achieving cost efficiency in these dynamic environments. However, traditional VM allocation and scheduling approaches often fail to account for the fluctuating and unpredictable nature of VM utilization, leading to inefficiencies such as over- or under-provisioning of resources. High-quality interval prediction helps accurately capture uncertainty in cloud resource demand and supports cloud operators in efficient instance provisioning. As an effective and reliable framework for constructing prediction intervals (PIs), conformal prediction (CP) is used for mid- and long-term for…",
            "url": "https://arxiv.org/abs/2607.24773",
            "image": "",
            "published": "2026-07-29T04:00:00+00:00",
            "score": 86.44,
            "color": "#b09cff",
            "raw": {
                "title": "Right-sizing Recommendations (RSR): Cloud Workload Conformal Prediction for Virtual Machines in Data Center Operations",
                "summary": "arXiv:2607.24773v1 Announce Type: new \nAbstract: Managing cloud infrastructure efficiently, especially in environments of large cloud providers or hyperscalers, requires optimizing the use of physical resources to minimize costs and maximize performance. Selecting the right virtual machine (VM) sizes is crucial to achieving cost efficiency in these dynamic environments. However, traditional VM allocation and scheduling approaches often fail to account for the fluctuating and unpredictable nature of VM utilization, leading to inefficiencies such as over- or under-provisioning of resources. High-quality interval prediction helps accurately capture uncertainty in cloud resource demand and supports cloud operators in efficient instance provisioning.\n  As an effective and reliable framework for constructing prediction intervals (PIs), conformal prediction (CP) is used for mid- and long-term forecasting tasks in cloud computing environments. This study proposes a new data-driven PI construction approach using bootstrapping conformal prediction for modern, dynamic, data-driven Right-sizing Recommendations (RSR) to enhance provisioning for diverse application workloads on hyperscalers. By learning workload utilization patterns, identifying correlations across multiple time series, and predicting medium- to long-term utilization trends, this research seeks to improve the efficiency of cloud and data center operations through an AI/ML-based provisioning pipeline.\n  Our study demonstrates that AI-driven models, powered by machine learning regression techniques and evaluated using backtesting, achieve promising forecasting results for cloud resource utilization. Additionally, we rank the selected models to identify top-performing approaches for long-life VM candidates. The proposed framework enhances right-sizing recommendations and supports more cost-effective resource allocation in dynamic cloud environments.",
                "url": "https://arxiv.org/abs/2607.24773",
                "image": "",
                "published": "Wed, 29 Jul 2026 00:00:00 -0400"
            }
        },
        {
            "id": "6a3cad09d84de7105ae9fad5d1152a2fc1ba8139",
            "source_id": "arxiv_ai",
            "source": "arXiv — Artificial Intelligence",
            "category": "ai",
            "engine": "rss",
            "title": "Atmospheric Diffusion-Guided Spatio-Temporal Transformer for Nuclear Radiation Forecasting",
            "summary": "arXiv:2607.24774v1 Announce Type: new Abstract: Nuclear radiation, the energy released during atomic decay, poses persistent risks to public health and the environment, and concerns have only grown since the Fukushima accident and the recent commencement of treated-water discharge. Modern monitoring networks now record radiation levels and accompanying weather conditions at thousands of stations, opening the door to nationwide forecasting that can inform emergency response, agricultural advisories, and routine public-safety decisions. However, turning this abundance of monitoring data into reliable forecasts is difficult for three reasons. First, the time series at each station are highly non-stationary, shaped by radioactive decay, weather variability, and irregular human interventions. Second, monitoring stations are severely unevenly distributed in space. Roughly 78% of Japan's stati…",
            "url": "https://arxiv.org/abs/2607.24774",
            "image": "",
            "published": "2026-07-29T04:00:00+00:00",
            "score": 86.44,
            "color": "#b09cff",
            "raw": {
                "title": "Atmospheric Diffusion-Guided Spatio-Temporal Transformer for Nuclear Radiation Forecasting",
                "summary": "arXiv:2607.24774v1 Announce Type: new \nAbstract: Nuclear radiation, the energy released during atomic decay, poses persistent risks to public health and the environment, and concerns have only grown since the Fukushima accident and the recent commencement of treated-water discharge. Modern monitoring networks now record radiation levels and accompanying weather conditions at thousands of stations, opening the door to nationwide forecasting that can inform emergency response, agricultural advisories, and routine public-safety decisions. However, turning this abundance of monitoring data into reliable forecasts is difficult for three reasons. First, the time series at each station are highly non-stationary, shaped by radioactive decay, weather variability, and irregular human interventions. Second, monitoring stations are severely unevenly distributed in space. Roughly 78% of Japan's stations sit in less than 6% of the country, clustered near Fukushima, which breaks the assumptions of standard graph-based models. Third, radiation co-evolves with heterogeneous context such as wind, temperature, and humidity through atmospheric transport processes that purely data-driven models struggle to capture from observations alone. In this study, we introduce NRFormer+, a spatio-temporal Transformer for nationwide nuclear radiation forecasting. NRFormer+ couples non-stationary temporal attention and density-adaptive spatial attention with a new atmospheric diffusion module that estimates how meteorology drives radiation dispersion and injects this physical signal into the network as an architectural prior. NRFormer+ delivers state-of-the-art accuracy on both datasets across all 13 baselines, reducing sudden-change MAE by up to 19.1% over the strongest baseline at comparable inference latency. Our code and datasets are publicly available at https://github.com/tfeilyu/NRFormer_Plus.",
                "url": "https://arxiv.org/abs/2607.24774",
                "image": "",
                "published": "Wed, 29 Jul 2026 00:00:00 -0400"
            }
        },
        {
            "id": "e4ff9190f2bc67338f7a7e61ec0979dbd2c6bad4",
            "source_id": "arxiv_ai",
            "source": "arXiv — Artificial Intelligence",
            "category": "ai",
            "engine": "rss",
            "title": "Steering topology distributions for unified generative design of architected metamaterials",
            "summary": "arXiv:2607.24777v1 Announce Type: new Abstract: Architected metamaterials derive their functions from structure, creating vast opportunities to program physical responses through topology design. However, existing design methods are often tailored to individual design problems, making limited use of topology knowledge for effective and broadly applicable design as objectives, constraints, and physical functions change. Here we introduce Generative Topology Optimization (GenTO), a unified framework that turns a learned topology prior into a reusable design engine. GenTO trains a diffusion model on a large full-order topology dataset and then iteratively steers the resulting topology distribution toward task-specific high-performing regions using user-defined physical objectives and constraints. This shifts the object of optimization from a single structure to a task-adapted topology dist…",
            "url": "https://arxiv.org/abs/2607.24777",
            "image": "",
            "published": "2026-07-29T04:00:00+00:00",
            "score": 86.44,
            "color": "#b09cff",
            "raw": {
                "title": "Steering topology distributions for unified generative design of architected metamaterials",
                "summary": "arXiv:2607.24777v1 Announce Type: new \nAbstract: Architected metamaterials derive their functions from structure, creating vast opportunities to program physical responses through topology design. However, existing design methods are often tailored to individual design problems, making limited use of topology knowledge for effective and broadly applicable design as objectives, constraints, and physical functions change. Here we introduce Generative Topology Optimization (GenTO), a unified framework that turns a learned topology prior into a reusable design engine. GenTO trains a diffusion model on a large full-order topology dataset and then iteratively steers the resulting topology distribution toward task-specific high-performing regions using user-defined physical objectives and constraints. This shifts the object of optimization from a single structure to a task-adapted topology distribution. Across topology design problems spanning thermal extremization, multi-objective morphology control, property-targeted auxetic design, and vibration transmission design, GenTO reuses pretrained topology priors for heterogeneous tasks, preserves structural diversity, and reaches high-performing solutions supported by numerical benchmarks and experimental validation. These results establish reusable topology knowledge as a unified principle for effective and scalable architected metamaterial design.",
                "url": "https://arxiv.org/abs/2607.24777",
                "image": "",
                "published": "Wed, 29 Jul 2026 00:00:00 -0400"
            }
        },
        {
            "id": "11d8523f3384c43bcb057f931d366a3a38f349ca",
            "source_id": "arxiv_ai",
            "source": "arXiv — Artificial Intelligence",
            "category": "ai",
            "engine": "rss",
            "title": "HOBA: Hierarchical On-Policy Bidding Agents for Adaptive Online Advertising",
            "summary": "arXiv:2607.24779v1 Announce Type: new Abstract: Online advertising bidding systems typically deploy multiple offline-trained expert models (e.g., PID controllers, model predictive control, offline RL policies) but face two critical limitations: lack of online adaptability to non-stationary auction markets, and reliance on costly manual tuning of hyperparameters such as bid bounds and budget pacing constraints. We propose HOBA (Hierarchical On-policy Bidding Agents), a hierarchical reinforcement learning framework that decouples strategic reasoning, model selection, and bid execution across three time scales. At the high level, a large language model infers hyperparameters from contextual signals through a Think-Act-Observe-Reflect loop with historical experience retrieval. At the mid level, a SARSA agent dynamically selects among expert models, incorporating causal adjustment to elimina…",
            "url": "https://arxiv.org/abs/2607.24779",
            "image": "",
            "published": "2026-07-29T04:00:00+00:00",
            "score": 86.44,
            "color": "#b09cff",
            "raw": {
                "title": "HOBA: Hierarchical On-Policy Bidding Agents for Adaptive Online Advertising",
                "summary": "arXiv:2607.24779v1 Announce Type: new \nAbstract: Online advertising bidding systems typically deploy multiple offline-trained expert models (e.g., PID controllers, model predictive control, offline RL policies) but face two critical limitations: lack of online adaptability to non-stationary auction markets, and reliance on costly manual tuning of hyperparameters such as bid bounds and budget pacing constraints. We propose HOBA (Hierarchical On-policy Bidding Agents), a hierarchical reinforcement learning framework that decouples strategic reasoning, model selection, and bid execution across three time scales. At the high level, a large language model infers hyperparameters from contextual signals through a Think-Act-Observe-Reflect loop with historical experience retrieval. At the mid level, a SARSA agent dynamically selects among expert models, incorporating causal adjustment to eliminate selection bias. At the low level, a dynamic expert pool (PID, MPC, IQL, Decision Transformer) executes bids under high-level constraints. This design confines online learning to discrete expert selection rather than continuous bid optimization, significantly reducing exploration risk while maintaining adaptability. Experiments on the AuctionNet benchmark and a large-scale A/B test demonstrate consistent improvements over state-of-the-art baselines. In a large-scale online deployment, HOBA delivered substantial business value, achieving a +3.6\\% increase in target cost, proving the effectiveness of our hierarchical multi-agent bidding paradigm.",
                "url": "https://arxiv.org/abs/2607.24779",
                "image": "",
                "published": "Wed, 29 Jul 2026 00:00:00 -0400"
            }
        },
        {
            "id": "d1c4f5e6b3ef905e366a337c91add23b51685e33",
            "source_id": "arxiv_ai",
            "source": "arXiv — Artificial Intelligence",
            "category": "ai",
            "engine": "rss",
            "title": "LivingArena: Do LLMs Know What Other LLMs Don't? Peer-Probing as Scalable Evaluation",
            "summary": "arXiv:2607.24780v1 Announce Type: new Abstract: Evaluating frontier LLMs is challenging: static benchmarks suffer from contamination and saturation -- leaving users unable to distinguish top models and developers blind to specific failure modes -- while human preference is subjective. In this paper, our question is: \\emph{Do LLMs know what other LLMs don't? And can we leverage this dynamic for evaluation?} We present \\textbf{LivingArena}, an automated, contamination-resistant evaluation framework. In this framework, models take turns proposing questions, aiming to pose items that opponents cannot answer correctly. Questioners are encouraged to actively identify and exploit opponents' knowledge boundaries, receiving rewards when the answerer fails, while the answerer is rewarded otherwise. To ensure questions contain objectively verifiable answers, a judge panel of strong models validate…",
            "url": "https://arxiv.org/abs/2607.24780",
            "image": "",
            "published": "2026-07-29T04:00:00+00:00",
            "score": 86.44,
            "color": "#b09cff",
            "raw": {
                "title": "LivingArena: Do LLMs Know What Other LLMs Don't? Peer-Probing as Scalable Evaluation",
                "summary": "arXiv:2607.24780v1 Announce Type: new \nAbstract: Evaluating frontier LLMs is challenging: static benchmarks suffer from contamination and saturation -- leaving users unable to distinguish top models and developers blind to specific failure modes -- while human preference is subjective. In this paper, our question is: \\emph{Do LLMs know what other LLMs don't? And can we leverage this dynamic for evaluation?} We present \\textbf{LivingArena}, an automated, contamination-resistant evaluation framework. In this framework, models take turns proposing questions, aiming to pose items that opponents cannot answer correctly. Questioners are encouraged to actively identify and exploit opponents' knowledge boundaries, receiving rewards when the answerer fails, while the answerer is rewarded otherwise. To ensure questions contain objectively verifiable answers, a judge panel of strong models validates them, penalizing questioners if validation fails. Evaluating ten frontier LLMs, LivingArena yields a stable Elo leaderboard. Our behavioral analyses show that models identify and exploit their peers' cognitive boundaries: self-play and tournament logs indicate that they localize and double down on opponents' weak dimensions. Beyond static knowledge recall, peer probing measures factual rigor and the higher-order ability to probe an opponent's weaknesses, correlating only weakly with human preference and offering a scalable, low-cost approach to continuous evaluation.",
                "url": "https://arxiv.org/abs/2607.24780",
                "image": "",
                "published": "Wed, 29 Jul 2026 00:00:00 -0400"
            }
        },
        {
            "id": "2535ed016a1bdb1835b753efc64f3d3a6554cc0f",
            "source_id": "arxiv_ai",
            "source": "arXiv — Artificial Intelligence",
            "category": "ai",
            "engine": "rss",
            "title": "Personalization, Personas, and Forecasting in Value Alignment",
            "summary": "arXiv:2607.24782v1 Announce Type: new Abstract: LLM behavior may be conditioned by human identity in several ways: they may be asked to adapt to users, role-play populations, or forecast how people would answer value-laden questions. We test whether these framings are interchangeable using the World Values Survey (WVS). We evaluate GPT-5.4, Claude Sonnet 4.6, Gemini 2.5 Flash, and Qwen3-235B on 101 WVS-derived questions across 13 language-country slices, comparing a language-only baseline with user-country, persona-country, and third-person prompts. Across 21,008 model-response rows, prompt framing is a first-order determinant of cultural alignment: country cues often shift answers substantially, but not all shifts move toward matched human response distributions. Third-person forecasting yields the strongest directional alignment for three of the four hosted models, while personalizati…",
            "url": "https://arxiv.org/abs/2607.24782",
            "image": "",
            "published": "2026-07-29T04:00:00+00:00",
            "score": 86.44,
            "color": "#b09cff",
            "raw": {
                "title": "Personalization, Personas, and Forecasting in Value Alignment",
                "summary": "arXiv:2607.24782v1 Announce Type: new \nAbstract: LLM behavior may be conditioned by human identity in several ways: they may be asked to adapt to users, role-play populations, or forecast how people would answer value-laden questions. We test whether these framings are interchangeable using the World Values Survey (WVS). We evaluate GPT-5.4, Claude Sonnet 4.6, Gemini 2.5 Flash, and Qwen3-235B on 101 WVS-derived questions across 13 language-country slices, comparing a language-only baseline with user-country, persona-country, and third-person prompts. Across 21,008 model-response rows, prompt framing is a first-order determinant of cultural alignment: country cues often shift answers substantially, but not all shifts move toward matched human response distributions. Third-person forecasting yields the strongest directional alignment for three of the four hosted models, while personalization and role-play are weaker or less stable. Alignment gains concentrate on salient value dimensions such as religiosity, gender roles, and work-oriented material values, whereas institutional trust and democracy-related questions remain difficult. These results show that prompt framing is not a cosmetic choice in cultural value elicitation; it changes both model behavior and measured alignment.",
                "url": "https://arxiv.org/abs/2607.24782",
                "image": "",
                "published": "Wed, 29 Jul 2026 00:00:00 -0400"
            }
        },
        {
            "id": "380c44958831ed0f19ceb68db4b0a8c1defb58b1",
            "source_id": "arxiv_ai",
            "source": "arXiv — Artificial Intelligence",
            "category": "ai",
            "engine": "rss",
            "title": "Unified Semantic Modeling Framework for Large-Scale Job Understanding at LinkedIn",
            "summary": "arXiv:2607.24783v1 Announce Type: new Abstract: Job understanding is critical to LinkedIn's mission of connecting talent with opportunity. This task involves transforming unstructured and noisy job postings into standardized or derived job attributes that power numerous LinkedIn products. However, building a scalable, cost-efficient, and high-performing job understanding system remains challenging. In this paper, we present a unified semantic modeling framework powered by a small language model (SLM) to address the challenges. We begin by fine-tuning an open-source SLM using a suite of carefully curated synthetic tasks augmented with reasoning traces. These tasks jointly target taxonomy-guided classification and taxonomy-agnostic entity extraction. This allows the resulting model to acquire robust zero-shot generalization for job understanding in structured and unstructured contexts. Bu…",
            "url": "https://arxiv.org/abs/2607.24783",
            "image": "",
            "published": "2026-07-29T04:00:00+00:00",
            "score": 86.44,
            "color": "#b09cff",
            "raw": {
                "title": "Unified Semantic Modeling Framework for Large-Scale Job Understanding at LinkedIn",
                "summary": "arXiv:2607.24783v1 Announce Type: new \nAbstract: Job understanding is critical to LinkedIn's mission of connecting talent with opportunity. This task involves transforming unstructured and noisy job postings into standardized or derived job attributes that power numerous LinkedIn products. However, building a scalable, cost-efficient, and high-performing job understanding system remains challenging. In this paper, we present a unified semantic modeling framework powered by a small language model (SLM) to address the challenges. We begin by fine-tuning an open-source SLM using a suite of carefully curated synthetic tasks augmented with reasoning traces. These tasks jointly target taxonomy-guided classification and taxonomy-agnostic entity extraction. This allows the resulting model to acquire robust zero-shot generalization for job understanding in structured and unstructured contexts. Building upon this foundation, we introduce a multi-adapter architecture with attribute grouping to facilitate efficient task-specific adaptation while streamlining model management across diverse downstream attributes. Offline evaluations and online A/B tests demonstrate significant performance improvement while reducing operational complexity. Our work provides practical insights into building industry-scale text understanding systems.",
                "url": "https://arxiv.org/abs/2607.24783",
                "image": "",
                "published": "Wed, 29 Jul 2026 00:00:00 -0400"
            }
        },
        {
            "id": "8991cffe5f088cb89833b286d9e3f946b0d2f4e9",
            "source_id": "arxiv_ai",
            "source": "arXiv — Artificial Intelligence",
            "category": "ai",
            "engine": "rss",
            "title": "On the Use of LLMs for Specialised Terminology: A Good Alternative to Corpora?",
            "summary": "arXiv:2607.24784v1 Announce Type: new Abstract: Specialised translation relies on the use of documentary and terminological resources, including corpora. These resources are particularly useful for terminology. However, their compilation and exploitation have several limitations: they require time, technical skills and access to data that can be difficult to collect. This study examines the extent to which LLMs can assist specialised translators in finding equivalents from English to French. We evaluate four proprietary models, GPT-4o, GPT-5.2, Claude Sonnet 4.5 and DeepSeek, in two specialised domains, Earth, Environmental and Planetary Sciences (EEPS) and Natural Language Processing (NLP). The experiment is based on 80 terms per domain and compares two prompting strategies: a terminology and a translation mode. The results highlight clear differences between models, prompting strategi…",
            "url": "https://arxiv.org/abs/2607.24784",
            "image": "",
            "published": "2026-07-29T04:00:00+00:00",
            "score": 86.44,
            "color": "#b09cff",
            "raw": {
                "title": "On the Use of LLMs for Specialised Terminology: A Good Alternative to Corpora?",
                "summary": "arXiv:2607.24784v1 Announce Type: new \nAbstract: Specialised translation relies on the use of documentary and terminological resources, including corpora. These resources are particularly useful for terminology. However, their compilation and exploitation have several limitations: they require time, technical skills and access to data that can be difficult to collect. This study examines the extent to which LLMs can assist specialised translators in finding equivalents from English to French. We evaluate four proprietary models, GPT-4o, GPT-5.2, Claude Sonnet 4.5 and DeepSeek, in two specialised domains, Earth, Environmental and Planetary Sciences (EEPS) and Natural Language Processing (NLP). The experiment is based on 80 terms per domain and compares two prompting strategies: a terminology and a translation mode. The results highlight clear differences between models, prompting strategies and, to a lesser extent, domains. Claude Sonnet 4.5 achieves the best results in the most favourable configuration, while DeepSeek stands out for its greater stability. Analysis of confidence estimates also shows that they are only a partial indicator of terminological accuracy. Overall, the findings suggest that LLMs can be useful tools for specialised translators, but cannot, at this stage, replace specialised corpora. This research therefore paves the way for future work on the real practical usefulness of LLMs for specialised translators in work and educational contexts.",
                "url": "https://arxiv.org/abs/2607.24784",
                "image": "",
                "published": "Wed, 29 Jul 2026 00:00:00 -0400"
            }
        },
        {
            "id": "dc7780f0252a0217acc1c27101292dbd5ebe589e",
            "source_id": "arxiv_ai",
            "source": "arXiv — Artificial Intelligence",
            "category": "ai",
            "engine": "rss",
            "title": "SpecPrefetch: Parameter-Efficient Expert Prefetching for Sparse MoE Foundation Models",
            "summary": "arXiv:2607.24787v1 Announce Type: new Abstract: Sparse Mixture-of-Experts (MoE) models expand foundation model capacity through conditional expert activation, but their full expert pools remain difficult to deploy under limited accelerator memory. Although expert offloading alleviates memory pressure by moving inactive experts to host memory or storage, it introduces a routing-dependent transfer bottleneck: required experts are known only after native top-\\(K\\) routing, which serializes routing, expert loading, and expert execution during inference. To address this bottleneck, we propose SpecPrefetch, a parameter-efficient prefetching framework for offloaded MoE inference. SpecPrefetch uses a shared lightweight adapter to predict next-layer expert candidates only for asynchronous transfer, while the frozen native router still determines the final executed experts. By separating transfer…",
            "url": "https://arxiv.org/abs/2607.24787",
            "image": "",
            "published": "2026-07-29T04:00:00+00:00",
            "score": 86.44,
            "color": "#b09cff",
            "raw": {
                "title": "SpecPrefetch: Parameter-Efficient Expert Prefetching for Sparse MoE Foundation Models",
                "summary": "arXiv:2607.24787v1 Announce Type: new \nAbstract: Sparse Mixture-of-Experts (MoE) models expand foundation model capacity through conditional expert activation, but their full expert pools remain difficult to deploy under limited accelerator memory. Although expert offloading alleviates memory pressure by moving inactive experts to host memory or storage, it introduces a routing-dependent transfer bottleneck: required experts are known only after native top-\\(K\\) routing, which serializes routing, expert loading, and expert execution during inference. To address this bottleneck, we propose SpecPrefetch, a parameter-efficient prefetching framework for offloaded MoE inference. SpecPrefetch uses a shared lightweight adapter to predict next-layer expert candidates only for asynchronous transfer, while the frozen native router still determines the final executed experts. By separating transfer prediction from execution routing, SpecPrefetch reduces exposed expert-loading latency without changing pretrained routing semantics, so prediction errors affect transfer efficiency rather than model outputs. In addition, a window-aware scheduler prioritizes feasible transfers under cache and bandwidth constraints. Across Qwen3-VL-30B-A3B and DeepSeek-VL2-Tiny, SpecPrefetch achieves the best average expert recall in 9 out of 10 model-benchmark settings with substantially fewer trainable parameters than learned predictor baselines. On a Snapdragon 8 Elite device, SpecPrefetch further improves decoding throughput by up to \\(20\\%\\) over a compute-optimized offloading runtime, demonstrating practical benefits for storage-constrained MoE deployment. The code and model weights are available at https://github.com/wei390/SpecPrefetch.",
                "url": "https://arxiv.org/abs/2607.24787",
                "image": "",
                "published": "Wed, 29 Jul 2026 00:00:00 -0400"
            }
        }
    ],
    "stored": 0,
    "ai": {
        "enriched": false,
        "stored": 0
    },
    "health": [
        {
            "id": "arxiv_ai",
            "name": "arXiv — Artificial Intelligence",
            "engine": "rss",
            "category": "ai",
            "ok": true,
            "error": "",
            "ms": 235,
            "count": 20,
            "cached_at": "2026-07-29T05:09:14+00:00"
        }
    ]
}