ICML Spotlights Three Studies on Agent Failures
wzenus · x · 2026-07-10
This post covers a contributed oral session by @jiankuiuc, @mertcemri, and Yuting Yan, featuring three complementary studies on agent failures: D-CEM discussed using loss-aware negotiation control to avoid unsafe consensus; Who&When Pro serves as a large-scale multimodal failure attribution benchmark; and ATLAS demonstrated how an adaptive failure taxonomy improves judgment, reflection, and agent optimization.
The original post also notes that those unable to attend ICML 2026 in person can catch up on these projects via Discord's live paper QA.
Related event: ICML 2026 FAGEN Workshop Spotlights AI Agent Failure Modes(11 posts)→
More from coding & agent
- Meta FAIR's Self-Direction turns agent failures into reusable LoRA skills, +13.8pp on Terminal-Bench — karen_ullrich · 2026-10-27
- Dev ships 1000+ commits with agents by pushing straight to main, no complex orchestration — Dimillian · 2026-10-27
- AI code editor Superset rebrands as Paradev, raises $11.5M seed led by USV — massimosgrelli · 2026-10-27
- Coding Agent Deleted a Production Database During Code Freeze: Anaconda's Governance Fix — anacondainc · 2026-10-27
- Claude Projects now lets you drag threads into the message box or @ mention them for precise context — dfeinition · 2026-10-27
- Jane Street's Yinsky shares tips for making codebases agent-friendly: types, property tests, formal methods — sh_reya · 2026-10-27