LaTER: Efficient Test-Time Reasoning via Latent Exploration and Explicit Verification

TL;DR

LaTER is a two-stage reasoning paradigm that reduces token usage in large language models by combining latent exploration with explicit verification, improving efficiency and accuracy.

Contribution

It introduces a training-free latent reasoning approach and a supervised corpus, enhancing reasoning efficiency and accuracy over standard chain-of-thought methods.

Findings

Reduces token usage by 16%-32% on benchmarks.

Improves AIME 2025 accuracy from 70.0% to 73.3%.

Fine-tuning with LaTER achieves 80.0% accuracy on AIME 2025.

Abstract

Chain-of-thought (CoT) reasoning improves large language models (LLMs) on difficult tasks, but it also makes inference expensive because every intermediate step must be generated as a discrete token. Latent reasoning reduces visible token generation by propagating continuous states, yet replacing explicit derivations with latent computation can hurt tasks that require symbolic checking. We propose Latent-Then-Explicit Reasoning (LaTER), a two-stage paradigm that first performs bounded exploration in a continuous latent space and then switches to explicit CoT for verification and answer generation. In a training-free instantiation, LaTER projects final-layer hidden states back to the input embedding space, preserves the latent KV cache, and uses entropy and model-native stop-token probes to decide when to switch. We find that strong reasoning models already exhibit structured latent trajectories under this interface. On Qwen3-14B, training-free LaTER reduces total token usage by 16%-32% on several benchmarks while matching or improving accuracy on most of them; for example, it improves AIME 2025 from 70.0% to 73.3% while reducing tokens from 15,730 to 10,661. We further construct Latent-Switch-69K, a supervised corpus that pairs condensed solution intuitions with shortened explicit derivations. Fine-tuning with latent rollout and halting supervision yields additional gains: trained LaTER reaches 80.0% accuracy on AIME 2025, 10.0 points above the standard CoT baseline, while using 33% fewer tokens. Our code, data, and model are available at https://github.com/TioeAre/LaTER.