LatentChem: From Textual CoT to Latent Thinking in Chemical Reasoning

TL;DR

LatentChem introduces a latent reasoning approach for chemical reasoning, enabling models to perform multi-step reasoning in continuous space, improving efficiency and performance over traditional text-based methods.

Contribution

It proposes a novel latent reasoning interface that decouples chemical computation from textual output, allowing implicit reasoning in continuous latent space.

Findings

Achieves a 59.88% non-tie win rate over CoT baselines on ChemCoTBench.

Reduces reasoning overhead by an average of 10.84 times.

Models internalize reasoning without explicit textual derivations.

Abstract

Chemical large language models (LLMs) predominantly rely on explicit Chain-of-Thought (CoT) in natural language to perform complex reasoning. However, chemical reasoning is inherently continuous and structural, and forcing it into discrete linguistic tokens introduces a fundamental representation mismatch that constrains both efficiency and performance. We introduce LatentChem, a latent reasoning interface that decouples chemical computation from textual generation, enabling models to perform multi-step reasoning directly in continuous latent space while emitting language only for final outputs. Remarkably, we observe a consistent emergent behavior: when optimized solely for task success, models spontaneously internalize reasoning, progressively abandoning verbose textual derivations in favor of implicit latent computation. This shift is not merely stylistic but computationally advantageous. Across diverse chemical reasoning benchmarks, LatentChem achieves a 59.88\% non-tie win rate over strong CoT-based baselines on ChemCoTBench, while delivering a 10.84$\times$ average reduction in reasoning overhead. Our results provide empirical evidence that chemical reasoning is more naturally and effectively realized as continuous latent dynamics rather than discretized linguistic trajectories.