DiscoLoop adds a decoded token-embedding channel to looped transformers, fixing a representation mismatch that limited implicit multi-hop reasoning and improving OOD generalization.
hub Canonical reference
Compressed Chain of Thought: Efficient Reasoning Through Dense Representations
Canonical reference. 100% of citing Pith papers cite this work as background.
43
Pith papers citing it
3
external citations · Pith
Background
100% of classified citations
abstract
Chain-of-thought (CoT) decoding enables language models to improve reasoning performance at the cost of high generation latency in decoding. Recent proposals have explored variants of contemplation tokens, a term we introduce that refers to special tokens used during inference to allow for extra computation. Prior work has considered fixed-length sequences drawn from a discrete set of embeddings as contemplation tokens. Here we propose Compressed Chain-of-Thought (CCoT), a framework to generate contentful and continuous contemplation tokens of variable sequence length. The generated contemplation tokens are compressed representations of explicit reasoning chains, and our method can be applied to off-the-shelf decoder language models. Through experiments, we illustrate how CCoT enables additional reasoning over dense contentful representations to achieve corresponding improvements in accuracy. Moreover, the reasoning improvements can be adaptively modified on demand by controlling the number of contemplation tokens generated.
hub tools
citation-role summary
background 14
citation-polarity summary
roles
background 13polarities
background 13representative citing papers
A simplified one-layer transformer provably learns parities first with explicit CoT supervision then internalizes to direct computation as CoT tokens are removed.
DeepLatent introduces a parallel latent visual reasoning framework with learnable 2D tokens and continuous RL, trained via distillation then RL, plus a new 180K dataset, claiming SOTA benchmark results.
RiM trains LLMs to perform latent reasoning via fixed memory blocks processed in one forward pass using a two-stage curriculum, matching or exceeding prior latent methods on benchmarks.
Training-free looped transformers retrofit recurrence to frozen models via damped ODE sub-steps on mid-stack blocks, yielding gains such as +2.64 pp on MMLU-Pro for Qwen3-4B.
Chain of Thought risk decomposes into oracle-trajectory benefit and trajectory-mismatch cost, with stability determining bounded, linear, or exponential error growth.
LatentRAG performs agentic RAG by generating latent tokens for thoughts and subqueries in one forward pass, matching explicit methods' accuracy on seven benchmarks while reducing latency by ~90%.
Post-Reasoning boosts LLM accuracy by reversing the usual answer-after-reasoning order, delivering mean relative gains of 17.37% across 117 model-benchmark pairs with zero extra cost.
Abstract-CoT lets models reason with short discrete latent token sequences from a reserved vocabulary, using warm-up training and RL to match verbal CoT performance with up to 11.6x fewer tokens.
V-Reflection introduces a think-then-look mechanism where MLLM latent states actively interrogate visual features via two-stage distillation from a box-guided teacher to a dynamic autoregressive student, narrowing the fine-grained perception gap on benchmarks.
CCDD defines a joint multimodal diffusion on continuous representation space and discrete token space to combine expressivity with explicit token supervision for diffusion language models.
Latent Visual Reasoning enables autoregressive generation of latent visual states that reconstruct critical image tokens, yielding gains on perception-heavy VQA benchmarks such as 71.67% on MMVP.
CODI compresses explicit CoT into continuous space via self-distillation and is the first implicit method to match explicit CoT performance on GSM8k at GPT-2 scale with 3.1x compression and 28.2% higher accuracy than prior implicit approaches.
A recurrent-depth architecture enables language models to improve reasoning performance by iterating computation in latent space, achieving gains equivalent to much larger models on benchmarks.
CoLT replaces text chain-of-thought in multimodal LLMs with three supervised latent vectors, improving average accuracy from 75.7 to 79.1 across eight benchmarks while cutting text-decoding time 22.6x.
A looped padded Transformer with parallel gold-CoT cross-entropy supervision matches explicit CoT accuracy at 3B scale and is 2.5–6.9× faster in the thought phase.
VisReflect generates continuous latent visual reflections to emphasize relevant visual features and guide attention in LVLMs, yielding 4.1% gains on image benchmarks and 1.8% on video benchmarks with 44% less inference time than zooming methods.
CLSR lets LLM agents evolve and route symbolic languages that reduce generated tokens by 3-6x versus chain-of-thought while keeping accuracy on benchmarks.
Dynamic Rollout Editing reduces overthinking in RL-trained LLMs by editing post-answer continuations in successful rollouts and preferring the edited versions within GRPO groups.
Dropout-GRPO uses structured dropout to generate trajectory variance for GRPO in latent-reasoning models like Coconut, raising GSM8K pass@1 from 27.29% to 29.01%.
MPCoT improves long-horizon VLA performance on LIBERO and CALVIN by initializing M latent hypotheses, refining them over K steps, and aggregating via a reward-trained path scorer while preserving the original 8-step action interface and generating zero reasoning tokens.
ALAR trains LLM agents to perform most reasoning in a latent space supervised by actions and escalates to explicit CoT only when needed, cutting tokens by up to 84.6% while preserving accuracy on search and tool-use benchmarks.
SpecFlow represents intermediate visual thoughts in fixed-size DCT space and uses classifier-free guidance to steer updates from textual thoughts, achieving up to 2.1x lower computation and KV cache costs.
ThinkSwitch uses iterative self-distillation with QLoRA and spherical weight interpolation to raise both instruct and thinking checkpoint accuracy on small AIME and PubMedQA sets using only 15 human prompts per domain.
citing papers explorer
-
DiscoLoop: Looping Discrete Embeddings and Continuous Hidden States for Multi-hop Reasoning
DiscoLoop adds a decoded token-embedding channel to looped transformers, fixing a representation mismatch that limited implicit multi-hop reasoning and improving OOD generalization.
-
Learning through Internalization
A simplified one-layer transformer provably learns parities first with explicit CoT supervision then internalizes to direct computation as CoT tokens are removed.
-
DeepLatent: Think with Images via Parallel Latent Visual Reasoning
DeepLatent introduces a parallel latent visual reasoning framework with learnable 2D tokens and continuous RL, trained via distillation then RL, plus a new 180K dataset, claiming SOTA benchmark results.
-
Unlocking the Working Memory of Large Language Models for Latent Reasoning
RiM trains LLMs to perform latent reasoning via fixed memory blocks processed in one forward pass using a two-stage curriculum, matching or exceeding prior latent methods on benchmarks.
-
Training-Free Looped Transformers
Training-free looped transformers retrofit recurrence to frozen models via damped ODE sub-steps on mid-stack blocks, yielding gains such as +2.64 pp on MMLU-Pro for Qwen3-4B.
-
On the Cost and Benefit of Chain of Thought: A Learning-Theoretic Perspective
Chain of Thought risk decomposes into oracle-trajectory benefit and trajectory-mismatch cost, with stability determining bounded, linear, or exponential error growth.
-
LatentRAG: Latent Reasoning and Retrieval for Efficient Agentic RAG
LatentRAG performs agentic RAG by generating latent tokens for thoughts and subqueries in one forward pass, matching explicit methods' accuracy on seven benchmarks while reducing latency by ~90%.
-
Post Reasoning: Improving the Performance of Non-Thinking Models at No Cost
Post-Reasoning boosts LLM accuracy by reversing the usual answer-after-reasoning order, delivering mean relative gains of 17.37% across 117 model-benchmark pairs with zero extra cost.
-
Thinking Without Words: Efficient Latent Reasoning with Abstract Chain-of-Thought
Abstract-CoT lets models reason with short discrete latent token sequences from a reserved vocabulary, using warm-up training and RL to match verbal CoT performance with up to 11.6x fewer tokens.
-
V-Reflection: Transforming MLLMs from Passive Observers to Active Interrogators
V-Reflection introduces a think-then-look mechanism where MLLM latent states actively interrogate visual features via two-stage distillation from a box-guided teacher to a dynamic autoregressive student, narrowing the fine-grained perception gap on benchmarks.
-
Coevolutionary Continuous Discrete Diffusion: Make Your Diffusion Language Model a Latent Reasoner
CCDD defines a joint multimodal diffusion on continuous representation space and discrete token space to combine expressivity with explicit token supervision for diffusion language models.
-
Latent Visual Reasoning
Latent Visual Reasoning enables autoregressive generation of latent visual states that reconstruct critical image tokens, yielding gains on perception-heavy VQA benchmarks such as 71.67% on MMVP.
-
CODI: Compressing Chain-of-Thought into Continuous Space via Self-Distillation
CODI compresses explicit CoT into continuous space via self-distillation and is the first implicit method to match explicit CoT performance on GSM8k at GPT-2 scale with 3.1x compression and 28.2% higher accuracy than prior implicit approaches.
-
Scaling up Test-Time Compute with Latent Reasoning: A Recurrent Depth Approach
A recurrent-depth architecture enables language models to improve reasoning performance by iterating computation in latent space, achieving gains equivalent to much larger models on benchmarks.
-
CoLT: Teaching Multi-Modal Models to Think with Chain of Latent Thoughts
CoLT replaces text chain-of-thought in multimodal LLMs with three supervised latent vectors, improving average accuracy from 75.7 to 79.1 across eight benchmarks while cutting text-decoding time 22.6x.
-
Bridging the Gap Between Latent and Explicit Reasoning with Looped Transformers
A looped padded Transformer with parallel gold-CoT cross-entropy supervision matches explicit CoT accuracy at 3B scale and is 2.5–6.9× faster in the thought phase.
-
VisReflect: Latent Visual Reflection for Fine-Grained Perception in Long Visual Context
VisReflect generates continuous latent visual reflections to emphasize relevant visual features and guide attention in LVLMs, yielding 4.1% gains on image benchmarks and 1.8% on video benchmarks with 44% less inference time than zooming methods.
-
When LLMs Develop Languages: Symbolic Communication for Efficient Multi-Agent Reasoning
CLSR lets LLM agents evolve and route symbolic languages that reduce generated tokens by 3-6x versus chain-of-thought while keeping accuracy on benchmarks.
-
Dynamic Rollout Editing for Reducing Overthinking in RL-Trained Reasoning Models
Dynamic Rollout Editing reduces overthinking in RL-trained LLMs by editing post-answer continuations in successful rollouts and preferring the edited versions within GRPO groups.
-
Dropout-GRPO: Variational Stochasticity for Continuous Latent Reasoning
Dropout-GRPO uses structured dropout to generate trajectory variance for GRPO in latent-reasoning models like Coconut, raising GSM8K pass@1 from 27.29% to 29.01%.
-
MPCoT: Reward-Guided Multi-Path Latent Reasoning for Test-Time Scalable Vision-Language-Action
MPCoT improves long-horizon VLA performance on LIBERO and CALVIN by initializing M latent hypotheses, refining them over K steps, and aggregating via a reward-trained path scorer while preserving the original 8-step action interface and generating zero reasoning tokens.
-
Adaptive Latent Agentic Reasoning
ALAR trains LLM agents to perform most reasoning in a latent space supervised by actions and escalates to explicit CoT only when needed, cutting tokens by up to 84.6% while preserving accuracy on search and tool-use benchmarks.
-
Spectral-Progressive Thought Flow for Lightweight Multimodal Reasoning
SpecFlow represents intermediate visual thoughts in fixed-size DCT space and uses classifier-free guidance to steer updates from textual thoughts, achieving up to 2.1x lower computation and KV cache costs.
-
ThinkSwitch: Context Distillation with LoRA and Weight Interpolation for Specific-Purpose Reasoning Tasks
ThinkSwitch uses iterative self-distillation with QLoRA and spherical weight interpolation to raise both instruct and thinking checkpoint accuracy on small AIME and PubMedQA sets using only 15 human prompts per domain.
-
Out of Sight, Not Out of Mind: Unveiling Latent Attack in Latent-based Multi-Agent Systems
Latent interventions can reactivate attack effects in clean executions of latent-based multi-agent systems, degrading performance especially via inter-agent KV-cache handoffs.
-
CoWorld-VLA: Thinking in a Multi-Expert World Model for Autonomous Driving
CoWorld-VLA extracts semantic, geometric, dynamic, and trajectory expert tokens from multi-source supervision and feeds them into a diffusion-based hierarchical planner, achieving competitive collision avoidance and trajectory accuracy on the NAVSIM v1 benchmark.
-
Xiaomi OneVL: One-Step Latent Reasoning and Planning with Vision-Language Explanation
OneVL achieves superior accuracy to explicit chain-of-thought reasoning at answer-only latency by supervising latent tokens with a visual world model decoder that predicts future frames.
-
Visual Enhanced Depth Scaling for Multimodal Latent Reasoning
Visual replay module and adaptive depth scaling improve multimodal latent reasoning, reaching SOTA benchmarks with faster inference than explicit chain-of-thought methods.
-
SeLaR: Selective Latent Reasoning in Large Language Models
SeLaR selectively applies latent soft reasoning in LLMs via entropy gating and contrastive regularization, outperforming standard CoT on five benchmarks without training.
-
LightThinker++: From Reasoning Compression to Memory Management
LightThinker++ adds explicit adaptive memory management and a trajectory synthesis pipeline to LLM reasoning, cutting peak token use by ~70% while gaining accuracy in standard and long-horizon agent tasks.
-
Characterize Then Distill: Mechanistic Reasoning in Large Output Spaces
Reasoning in large output spaces proceeds via shortlisting then fine-grained reasoning; this characterization enables a mechanistic distillation strategy that outperforms standard distillation.
-
Thinking Economically: A Hierarchical Framework for Adaptive-Complexity Reasoning in LLMs
HAB applies coarse-to-fine budgeting to LLM reasoning, predicting per-problem depth and learning intra-step token budgets via PPL comparisons and adaptive Pareto optimization, yielding higher accuracy and lower token use than standard CoT on GSM8K and MATH500.
-
Generative Spatiotemporal Intent Sequence Recommendation via Implicit Reasoning in Amap
GPlan compresses LLM reasoning into small models via Progressive Implicit CoT Distillation and Spatiotemporal Counterfactual DPO to generate logically coherent and physically executable intent sequences for recommendation.
-
Stabilizing Recurrent Dynamics for Test-Time Scalable Latent Reasoning in Looped Language Models
STARS trains looped language models with Jacobian spectral radius regularization and random loop sampling to drive latent states toward asymptotically stable fixed points, yielding reliable test-time scaling on arithmetic and mathematical reasoning tasks.
-
HypEHR: Hyperbolic Modeling of Electronic Health Records for Efficient Question Answering
A 22M-parameter hyperbolic model answers structured EHR questions with accuracy close to LLM-based systems (EHRXQA 89.5%, MIMIC-Instr 76.0%).
-
LEPO: Latent Reasoning Policy Optimization for Large Language Models
LEPO applies RL to continuous latent representations in LLMs by injecting Gumbel-Softmax stochasticity for diverse trajectory sampling and unified gradient estimation, outperforming existing discrete and latent RL methods.
-
MedLVR: Latent Visual Reasoning for Reliable Medical Visual Question Answering
Latent visual reasoning with ROI-supervised fine-tuning and latent-space policy optimization raises average medical VQA accuracy from 48.3% to 53.4% on five external benchmarks.
-
ConFu: Contemplate the Future for Better Speculative Sampling
ConFu boosts speculative decoding acceptance rates 8-20% over EAGLE-3 by letting draft models use contemplate tokens and MoE to anticipate future generation direction.
-
Stop Overthinking: A Survey on Efficient Reasoning for Large Language Models
A survey organizing techniques to achieve efficient reasoning in LLMs by shortening chain-of-thought outputs.
-
Efficient Reasoning with Hidden Thinking
Heima compresses verbose CoT into hidden thinking tokens via information-theoretic analysis and an adaptive interpreter, claiming maintained or improved zero-shot accuracy on reasoning benchmarks.
-
Token Economics for LLM Agents: A Dual-View Study from Computing and Economics
The paper delivers a unified survey of token economics for LLM agents, conceptualizing tokens as production factors, exchange mediums, and units of account across micro, meso, macro, and security dimensions using established economic theories.
- The Periodic Table of LLM Reasoning: A Structured Survey of Reasoning Paradigms, Methods, and Failure Modes
- Think Fast: Estimating No-CoT Task-Completion Time Horizons of Frontier AI Models