TSR · The desk
Science · 3 October 2026
Articles in this category in 3 October 2026.
TSR Desk · science · 3 October 2026, 01:00 UTC
AVSD-Scenes: A Dataset for Audio-Visual Description of Urban ScenesTSR Desk · science · 3 October 2026, 01:00 UTC
Pay for the Fault, Not the Flow: Label-Free In-Flow Multi-Agent Workflow OptimizationTSR Desk · science · 3 October 2026, 01:00 UTC
Hob-VL: A Benchmark for Visually Grounded Boolean ReasoningTSR Desk · science · 3 October 2026, 01:00 UTC
Where's Waldo? Query-language Preference under Cross-lingual Knowledge DisparitiesTSR Desk · science · 3 October 2026, 01:00 UTC
What Should an Agent Remember? Disentangling Retention from Retrieval in Bounded-MemoryTSR Desk · science · 3 October 2026, 01:00 UTC
Measuring the Stability Assumption Behind Action ChunkingTSR Desk · science · 3 October 2026, 01:00 UTC
The Hitchhikers Guide to Rubric Quality Understanding and EnrichmentTSR Desk · science · 3 October 2026, 01:00 UTC
Four Ways to Grow a Classifier and Why One of Them Cannot LearnTSR Desk · science · 3 October 2026, 01:00 UTC
Conflicting Supervision Moves Commitment, Not Capability: A 12.29{\sigma} arrangement effectTSR Desk · science · 3 October 2026, 01:00 UTC
Finding the Right Fit: Model-Harness Interactions across Agent TasksTSR Desk · science · 3 October 2026, 01:00 UTC
ROGUE: Evaluating Corrigibility Failures in Frontier Computer-Use AgentsTSR Desk · science · 3 October 2026, 01:00 UTC
OpenMTB-Audit: Exposing Over-Refusal and Clinical Expert Perspectives in LLM-Based MolecularTSR Desk · science · 3 October 2026, 01:00 UTC
When Does a Second Model Help? Cross-Model Review in LLM VerificationTSR Desk · science · 3 October 2026, 01:00 UTC
Pepti-drift: Scalable Safe-Active Peptide Generation Without Inference-Time GuidanceTSR Desk · science · 3 October 2026, 01:00 UTC
Scientific Agents: Evaluating Profession-Specific System Prompts on Scientific TasksTSR Desk · science · 3 October 2026, 01:00 UTC
How Divergence Becomes Decision Flips in Compressed Language ModelsTSR Desk · science · 3 October 2026, 01:00 UTC
Auto-Formalizing Neuro-Symbolic PredictorsTSR Desk · science · 3 October 2026, 01:00 UTC
FuncBridge: Towards Functional Tool-Use Generalization via Keypoint Trajectory ReasoningTSR Desk · science · 3 October 2026, 01:00 UTC
Q-Learning for Reachability in MEC-Free MDPs