Harness Handbook: Making Evolving Agent Harnesses Readable,Navigable, and Editable Paper • 2607.13285 • Published Jul 14 • 236
Blind-Spots-Bench: Evaluating Blind Spots in Multimodal Models Paper • 2607.08317 • Published Jul 9 • 38
ScarfBench: A Benchmark for Cross-Framework Application Migration in Enterprise Java Paper • 2605.06754 • Published May 18 • 1
HARC: Coupling Harmfulness and Refusal Directions for Robust Safety Alignment Paper • 2607.00572 • Published Jul 1 • 4
FastContext: Training Efficient Repository Explorer for Coding Agents Paper • 2606.14066 • Published Jun 12 • 96
SWE-Explore: Benchmarking How Coding Agents Explore Repositories Paper • 2606.07297 • Published Jun 5 • 124
GrepSeek: Training Search Agents for Direct Corpus Interaction Paper • 2605.29307 • Published May 28 • 118
Code2LoRA: Hypernetwork-Generated Adapters for Code Language Models under Software Evolution Paper • 2606.06492 • Published Jun 4 • 96
A Local Perturbation Theory for Cross-Domain Interference and Recovery in Multi-Domain RL Paper • 2606.02398 • Published Jun 1 • 28
Beyond Semantic Similarity: Rethinking Retrieval for Agentic Search via Direct Corpus Interaction Paper • 2605.05242 • Published May 3 • 127
When Models Manipulate Manifolds: The Geometry of a Counting Task Paper • 2601.04480 • Published Jan 8 • 4
Step 3.5 Flash: Open Frontier-Level Intelligence with 11B Active Parameters Paper • 2602.10604 • Published Feb 11 • 202
OCRVerse: Towards Holistic OCR in End-to-End Vision-Language Models Paper • 2601.21639 • Published Jan 29 • 52
Advances and Frontiers of LLM-based Issue Resolution in Software Engineering: A Comprehensive Survey Paper • 2601.11655 • Published Jan 15 • 63