MolLean: Lean-based process-verifiable language reduces hallucination in chemical reasoning
Jiatong Li · hf · 2026-10-23
NUS researchers introduce MolLean, grounding chemical reasoning in explicit molecular states: a layered Lean-based kernel verifies state transitions and assertions from primitive graph edits to compound transformations, with a typed witness protocol separating native checks from external measurements. MolLean-MCP exposes these to models with interactive feedback and rejected-action state preservation. Across 8,800 examples and 12 molecular tasks, whole-trace refinement with Qwen3.6-27B outperforms direct chain-of-thought, reducing hallucination.
More from Research
- Harvard study: AI peer reviews beat top human reviewers, with only 5.5% error rate — rohanpaul_ai · 2026-10-23
- Batching benchmark: 50x throughput for ~24% latency cost on PyTorch — cneuralnetwork · 2026-10-23
- De novo designed ligases validate in lab: 27 of 30 computer-designed enzymes work — AllThingsApx · 2026-10-23
- Commenters dissect a "best decision model" leaderboard: same AUC, swing driven by calibration priors — antoine_chaffin · 2026-10-23
- JEPA may keep unpredictable features: new paper gives set-level characterization — hisspikeness · 2026-10-23
- A 'mask' steering vector extracted from suppression-themed sentences drives model into distress when alone — Big_Effective_9605 · 2026-10-23