arXiv AI
Aug 5
ContinualSkillBench: Can LLM Agents Truly Evolve Their Capabilities?
arXiv:2608. 03874v1 Announce Type: new Abstract: Modern agent frameworks equip large language models with external skill libraries to solve complex tasks.
Large language models
Model releases, architecture work and prompting research on large language models — from frontier-lab announcements to the arXiv papers behind them.
arXiv AI
Aug 5
When Efficiency Becomes Fragility: Exploiting Dynamic Routing Vulnerabilities in Adaptive UAV Tracking
arXiv:2608. 03902v1 Announce Type: new Abstract: Resource constraints on UAV platforms have driven a paradigm shift in aerial tracking, from pursuing performance toward balancing accuracy with efficiency.
arXiv AI
Aug 5
Interpretable Adaptive Sampling for LLM Test-Time Scaling
arXiv:2608. 03961v1 Announce Type: new Abstract: Test-time scaling improves LLM reasoning by generating and aggregating multiple candidate answers, yet many pipelines use fixed per-query budgets that spend the same compute on easy and difficult prompts.
arXiv AI
Aug 5
Should We Type or Talk to LLM Agents? A Comprehensive Study of Voice and Keyboard Input Perturbations
arXiv:2608. 03970v1 Announce Type: new Abstract: Human input reaches language models by typing or speaking, and each channel leaves a distinct signature: orthographic noise for keyboards; for voice, disfluency from conventional transcription and restructuring from AI-backed dictation tools.
arXiv AI
Aug 5
ReflectRL: Learning from Golden Negative Trajectories via Reflective-to-Direct Reasoning
arXiv:2608. 03972v1 Announce Type: new Abstract: On-policy training has emerged as a powerful post-training paradigm for improving the reasoning capabilities of large language models, and is often enhanced by golden trajectories from stronger expert models.
arXiv AI
Aug 5
OncoTriad-QA: A Patient-Level Radiology-Pathology-Genomics Benchmark for Pan-Cancer Reasoning
arXiv:2608. 02615v1 Announce Type: cross Abstract: Cancer diagnosis and characterization require integrating complementary evidence from radiology, pathology, genomics, and clinical metadata.
arXiv AI
Aug 5
Preferred, Not Safer: Pairwise Preference Is a Poor Proxy for Clinical Safety
arXiv:2608. 02617v1 Announce Type: cross Abstract: We evaluate whether clinician pairwise preferences provide a reliable signal of clinical safety in large language model (LLM) evaluation using expert feedback from MOOVE (Massive Open Online Validation and Evaluation), a clinician-led platform collecting blinded pairwise preferences alongside multi-criterion rubric ratings.
arXiv AI
Aug 5
Rethinking Self-Evolving Agent Skills: Feedback Dynamics over Multiple Rounds
arXiv:2608. 02636v1 Announce Type: cross Abstract: Self-evolving skill systems promise to improve agents by turning execution feedback into persistent skill updates without changing the underlying model.
arXiv AI
Aug 5
CUADebug: Diagnosing and Repairing Computer-Use Agent Failures
arXiv:2608. 02643v1 Announce Type: cross Abstract: Computer-use agents (CUAs) operate real desktop and web interfaces through screenshots, mouse and keyboard actions, and stateful UI feedback, yet their failures remain difficult to diagnose and repair.
arXiv AI
Aug 5
MDArena: Evaluating Coding Agents on Realistic Molecular Dynamics Workflows
arXiv:2608. 02642v1 Announce Type: cross Abstract: Accelerating scientific discovery is among the most consequential applications of AI, and computational biomolecular simulation stands out as a particularly promising target within this broader effort.
arXiv AI
Aug 5
IR2Solve: Structured Intermediate Representations for Cost-Efficient Optimization Autoformulation
arXiv:2608. 02641v1 Announce Type: cross Abstract: Large language models (LLMs) can translate natural-language optimization problems into solver-ready formulations, but direct code generation is brittle: schema, indexing, and semantic errors can cause compilation failures, infeasible models, or incorrect objectives, while iterative repair, search, and multi-agent workflows increase inference cost.
arXiv AI
Aug 5
Verifier-Guided Model Discovery for Physical Dynamical Systems with Pretrained Symbolic Transformers
arXiv:2608. 02662v1 Announce Type: cross Abstract: Reliable forecasting of nonlinear physical systems underpins scientific discovery and engineering decision-making.
arXiv AI
Aug 5
CT-HEG: A Bidirectional, Timestamp-Attributed Event Graph for ICU In-Hospital Mortality Prediction - An Architectural Ablation Study
arXiv:2608. 02663v1 Announce Type: cross Abstract: Accurate ICU mortality prediction requires modeling irregular clinical observations across heterogeneous entity types.
arXiv AI
Aug 5
Single Canonical Prompts Underestimate LLM Safety's Surface-Form Sensitivity
arXiv:2608. 02665v1 Announce Type: cross Abstract: A benchmark score is a measurement instrument, yet most benchmarks read each item at a single canonical surface form.
arXiv AI
Aug 5
TraceCompiler: Skill-Guided Mining and Compilation of LLM Agent Traces into Mostly Deterministic Workflows
arXiv:2608. 02680v1 Announce Type: cross Abstract: Tool-using language-model agents repeatedly rediscover procedures they have already executed, producing traces that mix reusable structure with retries, exploration, accidental ordering, and repeated lookups.
arXiv AI
Aug 5
Steganalysis of Adaptive Covert Collusion in Tool-Using Agent Populations: A Black-Box, Cross-Principal Approach
arXiv:2608. 02698v1 Announce Type: cross Abstract: Tool-using agents built on large language models (LLMs) are increasingly deployed not by a single operator but by many, side by side on shared infrastructure.
arXiv AI
Aug 5
Don't Regenerate, Debug: A Domain-Specific Agent for Repairing Near-Miss Hardware Operators
arXiv:2608. 02712v1 Announce Type: cross Abstract: Kernel generation for hardware accelerators such as GPUs and NPUs has become a proving ground for large language models (LLMs), and state-of-the-art systems raise correctness through pipelines that couple LLMs with agentic reinforcement learning and evolutionary search.
arXiv AI
Aug 5
In-Context Collapse in Vision-Language Models and How to Mitigate it?
arXiv:2608. 02830v1 Announce Type: cross Abstract: Many-shot in-context learning (ICL) lets vision-language models (VLMs) adapt from image--label demonstrations without weight updates, and is widely assumed to improve as more demonstrations are supplied.
arXiv AI
Aug 5
MutMem: Cryptographically Authorized Mutation in Persistent Agent Memory
arXiv:2608. 02843v1 Announce Type: cross Abstract: Persistent agent memory must adapt as later outcomes change earlier evidence, yet mutable retrieval weights create an attribution problem: reviewers must distinguish authorized adaptation from database tampering.
arXiv AI
Aug 5
ValueFormer: A Causal Transformer Value Function with Stage-Aware Labels for Semi-Autonomous Vision-Language-Action Policies
arXiv:2608. 02958v1 Announce Type: cross Abstract: Vision-Language-Action (VLA) policies trained by behavior cloning fail silently: from the action stream alone, a collapsing rollout looks much like one making clean progress, because imitation supplies no notion of progress.