Two EMNLP mech-interp papers: LLM agents as circuit explainers and DuoSteer for safe code generation
ZiyuYao · x · 2026-10-26
Two mech-interp posters are coming to EMNLP in Budapest: (1) 'Can Language Model Agents be Helpful Circuit Explainers in Mechanistic Interpretability?', studying automated interpretability via LLM agents; (2) 'Interpreting and Steering for Safe and Correct Code Generation', which examines how code-vulnerability representations interfere with code correctness and proposes DuoSteer for steering. Posters run in Hall H on Oct 25 and Oct 27.
More from Research
- CoRL paper: one-step inference matches multi-step for large VLAs, 1.6-4.8x faster action generation — kaixhin · 2026-10-26
- Why biology is hard to simulate: universally sloppy parameter sensitivities, per classic PLOS paper — rbhar90 · 2026-10-26
- LessWrong post argues FDT is difficult-to-impossible to specify due to logical counterfactuals — jessi_cata · 2026-10-26
- Bitter Lesson misread: scaling won't brute-force biology, researcher argues — anshulkundaje · 2026-10-26
- Chain Theory: rebuilding physics from quantum information, AI finds 137 fine-structure-constant expressions — skdh · 2026-10-26
- HKU/ZJU/Fudan with Lightspeed launch HiSTEVE: hierarchical VLA agent playing Minecraft via native mouse and keyboard — tencent · 2026-10-26