12 posts published in the last hour
- 05:32DocHop: Benchmarking Out-of-domain Multi-hop Reasoning in Information-Dense Documents
- 05:32HeadWiseKV: Budgeted Per-Head Cache Residency for Hybrid Long-Context Language Models
- 05:32MineTRACE: An Evidence-Grounded Interactive Reasoning System for Mineral Prospectivity
- 05:32Anthropic’s Claude Fable 5.1 promises better coding and research at up to 45 percent less
- 05:32Monitoring Web Agents Without Internal Signals: Observable Trajectories and Key-Step Supervision
- 05:32Anthropic’s new Fable release is cheaper, less restrictive
- 05:32ClaimReceipt: Verifying Evidence Sufficiency and Coverage in Agent Evaluations
- 05:03Looped Transformers under the Jacobian Lens: Does the Global Workspace Survive Recurrence?
- 05:03When Agents Implement Systems: A Case Study in Defects, Detection, and Evaluation Rigor
- 05:02Benchmarking Language Models for Statistical Problem Formulation
- 05:02The Ceiling Is in the Channel: Auditing Learner Gaps and Measurement Frontiers in Clinical Prediction
- 05:02Post-Training Ternarization of Qwen3-4B Capability, Effective Bit Budget, Storage Compression, and Deployment