SLIM decomposes LLM hidden states via sparse autoencoders with learnable gates to enable precise, interpretable steering of molecular properties, yielding up to 42.4-point gains on the MolEditRL benchmark.
InProceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.LG 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
TRACE is a trajectory-aware LLM agent that treats molecular tool selection as sequential decision-making to achieve higher success rates and larger ADMET improvements than one-step baselines on optimization tasks.
citing papers explorer
-
SLIM: Sparse Latent Steering for Interpretable and Property-Directed LLM-Based Molecular Editing
SLIM decomposes LLM hidden states via sparse autoencoders with learnable gates to enable precise, interpretable steering of molecular properties, yielding up to 42.4-point gains on the MolEditRL benchmark.
-
Molecular Lead Optimization via Agentic Tool Planning
TRACE is a trajectory-aware LLM agent that treats molecular tool selection as sequential decision-making to achieve higher success rates and larger ADMET improvements than one-step baselines on optimization tasks.