12 posts published in the last hour
- 08:34KPHarness: Co-Evolving Knowledge and Probing for Long-Horizon Gaming Harnesses
- 08:34Ask to Distill: Learning to Synthesize Data for On-Policy Distillation
- 08:34Recoverability and Verifier Strength: Two Measured Preconditions for Unattended Agent Self-Improvement
- 08:33Runtime Control for AI Agents: Closing the Evaluation–Execution Gap in Operational Workflows
- 08:33OpenAI acquires Orchestra Bio to build AI agents that speed up drug development
- 08:33Harness Self-Evolution for Long-Horizon Terminal Agents
- 08:06Correcting Attribution Patching with a Few Exact Measurements
- 08:06Unified Sharp Bounds on Probabilities of Causation
- 08:06More Memory, Worse Agents: Error Reproduction and Anti-Persistence in LLM Agents
- 08:05When PTQ Calibration Helps or Hurts: Dose-Driven Collapse and Validation-Gated Abstention in Quantized Text Embedders
- 08:05ProMemAgent: Long-Horizon Proactive Assistance with Context-Aware Prospective Memory from Sensory Perceptions
- 08:00AI News Brief Hourly Summary 2026-10-21 10h : 18 posts