SeLaR: Selective Latent Reasoning in Large Language Models

• URL: http://arxiv.org/abs/2604.08299v1
• Date: Thu, 09 Apr 2026 14:32:07 GMT
• Title: SeLaR: Selective Latent Reasoning in Large Language Models
• Abstract summary: Chain-of-Thought (CoT) has become a cornerstone of reasoning in large language models.<n>Recent latent reasoning approaches attempt to alleviate this limitation by replacing discrete tokens with soft embeddings.<n>We propose SeLaR, a lightweight and training-free framework for latent reasoning.
• Score: 5.837876642209202
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Chain-of-Thought (CoT) has become a cornerstone of reasoning in large language models, yet its effectiveness is constrained by the limited expressiveness of discrete token sampling. Recent latent reasoning approaches attempt to alleviate this limitation by replacing discrete tokens with soft embeddings (probability-weighted mixtures of token embeddings) or hidden states, but they commonly suffer from two issues: (1) global activation injects perturbations into high-confidence steps, impairing reasoning stability; and (2) soft embeddings quickly collapse toward the highest-probability token, limiting exploration of alternative trajectories. To address these challenges, we propose SeLaR (Selective Latent Reasoning), a lightweight and training-free framework. SeLaR introduces an entropy-gated mechanism that activates soft embeddings only at low-confidence steps, while preserving discrete decoding at high-confidence steps. Additionally, we propose an entropy-aware contrastive regularization that pushes soft embeddings away from the dominant (highest-probability) token's direction, encouraging sustained exploration of multiple latent reasoning paths. Experiments on five reasoning benchmarks demonstrate that SeLaR consistently outperforms standard CoT and state-of-the-art training-free methods.

Related papers

• RuPLaR : Efficient Latent Compression of LLM Reasoning Chains with Rule-Based Priors From Multi-Step to One-Step [7.647393987025467]
  We introduce One-Model One-Step, a novel compression framework for Latent Reasoning with Rule-Based Priors.<n>Our method trains an LLM to autonomously generate latent reasoning tokens in a single training stage.<n>Experiments show that our framework improves accuracy by 11.1% over existing latent CoT methods.
  arXiv  Detail & Related papers  (2026-05-10T05:55:07Z)
• Latent-GRPO: Group Relative Policy Optimization for Latent Reasoning [56.60090698761805]
  Existing latent reasoning methods mainly focus on supervised learning, and reinforcement learning in latent space remains highly unstable.<n>We study this problem through the lens of Group Relative Policy Optimization (GRPO), and show that directly adapting GRPO to latent reasoning is fundamentally non-trivial.<n>We propose textbfLatent-GRPO, which combines invalid-sample advantage masking, one-sided noise sampling, and optimal correct-path first-token selection.
  arXiv  Detail & Related papers  (2026-04-30T15:23:40Z)
• SPOT: Span-level Pause-of-Thought for Efficient and Interpretable Latent Reasoning in Large Language Models [15.95627037350657]
  Explicit Chain-of-Thought (CoT) computation incurs high inference cost due to internalize token-level traces.<n>We propose SPOT, a flexible framework that compresses explicit CoT into compact latent pause tokens without enforcing a fixed response template.<n>Experiments on reasoning benchmarks demonstrate that SPOT improves accuracy by 2.3 points on average while reducing generated tokens by 37.5%.
  arXiv  Detail & Related papers  (2026-03-06T12:34:27Z)
• Latent Thoughts Tuning: Bridging Context and Reasoning with Fused Information in Latent Tokens [13.653741247835091]
  Latent Thoughts Tuning (LT-Tuning) is a framework that redefines how latent thoughts are constructed and deployed.<n>We introduce a Context-Prediction-Fusion mechanism that jointly leveraging contextual hidden states and predictive semantic guidance.<n>Our method outperforms existing latent reasoning baselines, effectively mitigating feature collapse and achieving robust reasoning accuracy.
  arXiv  Detail & Related papers  (2026-02-10T19:19:10Z)
• Latent-Space Contrastive Reinforcement Learning for Stable and Efficient LLM Reasoning [16.244366307890832]
  We propose textbfDeepLatent Reasoning (DLR), a latent-space bidirectional contrastive reinforcement learning framework.<n>This framework shifts the trial-and-error cost from expensive token-level full sequence generation to the continuous latent manifold.<n> Experiments demonstrate that DLR achieves more stable training convergence, supports longer-horizon reasoning chains, and facilitates the sustainable accumulation of reasoning capabilities.
  arXiv  Detail & Related papers  (2026-01-24T03:18:22Z)
• Efficient Thought Space Exploration through Strategic Intervention [54.35208611253168]
  We propose a novel Hint-Practice Reasoning (HPR) framework that operationalizes this insight through two synergistic components.<n>The framework's core innovation lies in Distributional Inconsistency Reduction (DIR), which dynamically identifies intervention points.<n> Experiments across arithmetic and commonsense reasoning benchmarks demonstrate HPR's state-of-the-art efficiency-accuracy tradeoffs.
  arXiv  Detail & Related papers  (2025-11-13T07:26:01Z)
• Think Consistently, Reason Efficiently: Energy-Based Calibration for Implicit Chain-of-Thought [33.267497114389734]
  Large Language Models (LLMs) have demonstrated strong reasoning capabilities through emphChain-of-Thought (CoT) prompting.<n>CoT methods rely on discrete token-level reasoning processes prone to error propagation and limited by vocabulary.<n>We propose EBM-CoT, an Energy-Based Chain-of-Thought framework that refines latent thought representations through an energy-based model.
  arXiv  Detail & Related papers  (2025-11-10T14:10:58Z)
• Latent Chain-of-Thought for Visual Reasoning [53.541579327424046]
  Chain-of-thought (CoT) reasoning is critical for improving the interpretability and reliability of Large Vision-Language Models (LVLMs)<n>We reformulate reasoning in LVLMs as posterior inference and propose a scalable training algorithm based on amortized variational inference.<n>We empirically demonstrate that the proposed method enhances the state-of-the-art LVLMs on seven reasoning benchmarks.
  arXiv  Detail & Related papers  (2025-10-27T23:10:06Z)
• Latent Reasoning in LLMs as a Vocabulary-Space Superposition [80.01651003144282]
  Large language models (LLMs) demonstrate strong reasoning abilities with chain-of-thought prompting, but explicit reasoning introduces substantial computational overhead.<n>Recent work on latent reasoning reduces this cost by reasoning in latent space without explicit supervision, but performance drops significantly.<n>To address this, we restrict the latent space to the column space of the LLM vocabulary, treating latent reasoning as a superposition over vocabulary probabilities.<n>Once latent reasoning concludes, it collapses into an eigenstate of explicit reasoning to yield the final answer.<n>Latent-SFT sets a new state of the art on GSM8k, matching explicit
  arXiv  Detail & Related papers  (2025-10-17T10:51:20Z)
• SwiReasoning: Switch-Thinking in Latent and Explicit for Pareto-Superior Reasoning LLMs [18.427352953242792]
  SwiReasoning is a training-free framework for large language models (LLMs) reasoning.<n>It switches between explicit and latent reasoning, guided by block-wise confidence estimated from entropy trends in next-token distributions.<n>It consistently improves average accuracy by 1.5%-2.8% across reasoning LLMs of different model families and scales.
  arXiv  Detail & Related papers  (2025-10-06T17:46:34Z)
• ConciseHint: Boosting Efficient Reasoning via Continuous Concise Hints during Generation [74.37307916314407]
  We propose a framework dubbed ConciseHint, which continuously encourages the reasoning model to speak concisely.<n>Experiments on the state-of-the-art LRMs, including DeepSeek-R1 and Qwen-3 series, demonstrate that our method can effectively produce concise reasoning.
  arXiv  Detail & Related papers  (2025-06-23T16:20:44Z)
• Soft Thinking: Unlocking the Reasoning Potential of LLMs in Continuous Concept Space [62.54887038032942]
  We introduce Soft Thinking, a training-free method that emulates human-like "soft" reasoning by generating soft, abstract concept tokens.<n>These concept tokens are created by the probability-weighted mixture of token embeddings, which form the continuous concept space.<n>In essence, each generated concept token encapsulates multiple meanings from related discrete tokens, implicitly exploring various reasoning paths to converge.
  arXiv  Detail & Related papers  (2025-05-21T17:29:15Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.