TokenSkip: Controllable Chain-of-Thought Compression in LLMs

• URL: http://arxiv.org/abs/2502.12067v2
• Date: Sat, 24 May 2025 01:34:51 GMT
• Title: TokenSkip: Controllable Chain-of-Thought Compression in LLMs
• Authors: Heming Xia, Chak Tou Leong, Wenjie Wang, Yongqi Li, Wenjie Li,
• Abstract summary: Chain-of-Thought (CoT) has been proven effective in enhancing the reasoning capabilities of large language models (LLMs)<n>We propose TokenSkip, a simple yet effective approach that enables LLMs to selectively skip less important tokens, allowing for controllable CoT compression.
• Score: 11.583847083770035
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Chain-of-Thought (CoT) has been proven effective in enhancing the reasoning capabilities of large language models (LLMs). Recent advancements, such as OpenAI's o1 and DeepSeek-R1, suggest that scaling up the length of CoT sequences during inference could further boost LLM reasoning performance. However, due to the autoregressive nature of LLM decoding, longer CoT outputs lead to a linear increase in inference latency, adversely affecting user experience, particularly when the CoT exceeds 10,000 tokens. To address this limitation, we analyze the semantic importance of tokens within CoT outputs and reveal that their contributions to reasoning vary. Building on this insight, we propose TokenSkip, a simple yet effective approach that enables LLMs to selectively skip less important tokens, allowing for controllable CoT compression. Extensive experiments across various models and tasks demonstrate the effectiveness of TokenSkip in reducing CoT token usage while preserving strong reasoning performance. Notably, when applied to Qwen2.5-14B-Instruct, TokenSkip reduces reasoning tokens by 40% (from 313 to 181) on GSM8K, with less than a 0.4% performance drop.

Related papers

• R-Stitch: Dynamic Trajectory Stitching for Efficient Reasoning [60.37610817226533]
  Chain-of-thought (CoT) reasoning encourages step-by-step intermediate reasoning during inference.<n>CoT introduces substantial computational overhead due to its reliance on autoregressive decoding over long token sequences.<n>We present R-Stitch, a token-level, confidence-based hybrid decoding framework that accelerates CoT inference.
  arXiv  Detail & Related papers  (2025-07-23T08:14:36Z)
• VeriThinker: Learning to Verify Makes Reasoning Model Efficient [52.74493506816969]
  Large Reasoning Models excel at complex tasks using Chain-of-Thought (CoT) reasoning.<n>Their tendency to overthinking leads to unnecessarily lengthy reasoning chains.<n>We introduce VeriThinker, a novel approach for CoT compression.
  arXiv  Detail & Related papers  (2025-05-23T14:17:56Z)
• Not All Tokens Are What You Need In Thinking [34.767739567093656]
  Conditional Token Selection (CTS) identifies and preserves only the most essential tokens in chains of thought.<n>CTS effectively compresses long CoT while maintaining strong reasoning performance.<n>Further reducing training tokens by 42% incurs only a marginal 5% accuracy drop while yielding a 75.8% reduction in reasoning tokens.
  arXiv  Detail & Related papers  (2025-05-23T12:41:29Z)
• R1-Compress: Long Chain-of-Thought Compression via Chunk Compression and Search [61.4807238517108]
  Chain-of-Thought (CoT) reasoning enhances large language models (LLMs) by enabling step-by-step problem-solving.<n>CoT's extension to Long-CoT introduces substantial computational overhead due to increased token length.<n>We propose R1-Compress, a two-stage chunk-level compression framework that preserves both local information and coherence.
  arXiv  Detail & Related papers  (2025-05-22T16:06:59Z)
• Think Silently, Think Fast: Dynamic Latent Compression of LLM Reasoning Chains [15.89404914539006]
  We introduce Compressed Latent Reasoning (CoLaR), a novel framework that dynamically compresses reasoning processes in latent space.<n>CoLaR achieves 14.1% higher accuracy than latent-based baseline methods at comparable compression ratios.<n>Our RL-enhanced CoLaR demonstrates performance gains of up to 5.4% while dramatically reducing latent reasoning chain length by 82.8%.
  arXiv  Detail & Related papers  (2025-05-22T11:40:26Z)
• Fractured Chain-of-Thought Reasoning [61.647243580650446]
  We introduce Fractured Sampling, a unified inference-time strategy that interpolates between full CoT and solution-only sampling.<n>We show that Fractured Sampling consistently achieves superior accuracy-cost trade-offs, yielding steep log-linear scaling gains in Pass@k versus token budget.
  arXiv  Detail & Related papers  (2025-05-19T11:30:41Z)
• CoT-Valve: Length-Compressible Chain-of-Thought Tuning [50.196317781229496]
  We introduce a new tuning and inference strategy named CoT-Valve, designed to allow models to generate reasoning chains of varying lengths.<n>We show that CoT-Valve successfully enables controllability and compressibility of the chain and shows better performance than the prompt-based control.
  arXiv  Detail & Related papers  (2025-02-13T18:52:36Z)
• LLMs Can Easily Learn to Reason from Demonstrations Structure, not content, is what matters! [53.84130385074551]
  Large reasoning models (LRMs) tackle complex reasoning problems by following long chain-of-thoughts (Long CoT) We find that a Large Language model (LLM) can effectively learn Long CoT reasoning through data-efficient supervised fine-tuning (SFT) and parameter-efficient low-rank adaptation (LoRA) With just 17k long CoT training samples, the Qwen2.5-32B-Instruct model achieves significant improvements on a wide range of math and coding benchmarks.
  arXiv  Detail & Related papers  (2025-02-11T08:48:48Z)
• SepLLM: Accelerate Large Language Models by Compressing One Segment into One Separator [65.62084602011596]
  Large Language Models (LLMs) have exhibited exceptional performance across a spectrum of natural language processing tasks.<n>We have identified a key pattern: certain seemingly meaningless special tokens (i.e., separators) contribute disproportionately to attention scores compared to semantically meaningful tokens.<n>We introduce SepLLM, a plug-and-play framework that accelerates inference by compressing these segments and eliminating redundant tokens.
  arXiv  Detail & Related papers  (2024-12-16T18:58:57Z)
• C3oT: Generating Shorter Chain-of-Thought without Compromising Effectiveness [18.073777359647515]
  Chain-of-Thought (CoT) before deriving the answer can improve the reasoning capabilities of large language models (LLMs)<n>However, the length of the generated CoT is much longer than the desired final answer, which results in additional decoding costs.<n>This paper presents a CoT compression framework that involves a compressor to compress an original longer CoT into a shorter CoT.
  arXiv  Detail & Related papers  (2024-12-16T11:12:45Z)
• Inference Optimal VLMs Need Only One Visual Token but Larger Models [54.01228554126122]
  Vision Language Models (VLMs) have demonstrated strong capabilities across various visual understanding and reasoning tasks. VLMs are often constrained by high latency during inference due to substantial compute required to process the large number of input tokens. We take some initial steps towards building approaches tailored for high token compression settings.
  arXiv  Detail & Related papers  (2024-11-05T18:54:21Z)
• From Sparse Dependence to Sparse Attention: Unveiling How Chain-of-Thought Enhances Transformer Sample Efficiency [17.612497960364916]
  Chain-of-thought (CoT) significantly enhances the reasoning performance of large language models (LLM) We demonstrate that CoT can substantially improve sample efficiency even when representation power is sufficient. We show that CoT simplifies the learning process by introducing sparse dependencies among input tokens, and leads to a sparse and interpretable attention.
  arXiv  Detail & Related papers  (2024-10-07T19:45:09Z)
• To CoT or not to CoT? Chain-of-thought helps mainly on math and symbolic reasoning [55.52872152909785]
  Chain-of-thought (CoT) via prompting is the de facto method for eliciting reasoning capabilities from large language models (LLMs) We show that CoT gives strong performance benefits primarily on tasks involving math or logic, with much smaller gains on other types of tasks.
  arXiv  Detail & Related papers  (2024-09-18T17:55:00Z)
• FFN-SkipLLM: A Hidden Gem for Autoregressive Decoding with Adaptive Feed Forward Skipping [49.66872823080736]
  Autoregressive Large Language Models (e.g., LLaMa, GPTs) are omnipresent achieving remarkable success in language understanding and generation. To mitigate overload incurred during generation, several early-exit and layer-dropping strategies have been proposed. We propose FFN-SkipLLM, which is an input-adaptive feed-forward skipping strategy.
  arXiv  Detail & Related papers  (2024-04-05T02:35:43Z)
• Can Separators Improve Chain-of-Thought Prompting? [10.398343318429367]
  Chain-of-thought (CoT) prompting is a simple and effective method for improving the reasoning capabilities of Large Language Models (LLMs) Inspired by human cognition, we introduce COT-SEP, a method that strategically employs separators at the end of each exemplar in CoT prompting.
  arXiv  Detail & Related papers  (2024-02-16T12:46:16Z)
• Chain-of-Thought Reasoning Without Prompting [40.92854235219315]
  CoT reasoning paths can be elicited from pre-trained language models by simply altering the textitdecoding process. The presence of a CoT in the decoding path correlates with a higher confidence in the model's decoded answer.
  arXiv  Detail & Related papers  (2024-02-15T18:55:41Z)
• Chain of Evidences and Evidence to Generate: Prompting for Context Grounded and Retrieval Augmented Reasoning [3.117335706912261]
  Chain of Evidences (CoE) and Evidence to Generate (E2G) are built upon two unique strategies. Instead of unverified reasoning claims, our innovative approaches leverage the power of "evidence for decision making" Our framework consistently achieves remarkable results across various knowledge-intensive reasoning and generation tasks.
  arXiv  Detail & Related papers  (2024-01-11T09:49:15Z)

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.