DSV-Mem: Evaluating Multimodal Memory in Professional Workflows for MLLM Agents Paper • 2610.08102 • Published Oct 6 • 1
Beyond the Teacher: Understanding and Preserving Distillation Overshoot in Agentic On-Policy Distillation Paper • 2610.25034 • Published 23 days ago • 1
DeLoop: Repeated Actions as Self-Supervision for Long-Horizon Agents Paper • 2610.19376 • Published 27 days ago • 22
CLASHarness: Benchmarking Multi-Agent Conflict in Computer Use via a Controllable and Verifiable Generation Pipeline Paper • 2610.19128 • Published 28 days ago • 34
TIR-Bench: A Comprehensive Benchmark for Agentic Thinking-with-Images Reasoning Paper • 2511.01833 • Published Nov 3, 2025 • 16