REVIEW 42 cited by
Reasoning Models Can Be Effective Without Thinking
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
Recent LLMs have significantly improved reasoning capabilities, primarily by including an explicit, lengthy Thinking process as part of generation. In this paper, we question whether this explicit thinking is necessary. Using the state-of-the-art DeepSeek-R1-Distill-Qwen, we find that bypassing the thinking process via simple prompting, denoted as NoThinking, can be surprisingly effective. When controlling for the number of tokens, NoThinking outperforms Thinking across a diverse set of seven challenging reasoning datasets--including mathematical problem solving, formal theorem proving, and coding--especially in low-budget settings, e.g., 51.3 vs. 28.9 on ACM 23 with 700 tokens. Notably, the performance of NoThinking becomes more competitive with pass@k as k increases. Building on this observation, we demonstrate that a parallel scaling approach that uses NoThinking to generate N outputs independently and aggregates them is highly effective. For aggregation, we use task-specific verifiers when available, or we apply simple best-of-N strategies such as confidence-based selection. Our method outperforms a range of baselines with similar latency using Thinking, and is comparable to Thinking with significantly longer latency (up to 9x). Together, our research encourages a reconsideration of the necessity of lengthy thinking processes, while also establishing a competitive reference for achieving strong reasoning performance in low-budget settings or at low latency using parallel scaling.
Forward citations
Cited by 42 Pith papers
-
Is Long-to-Short a Free Lunch? Investigating Inconsistency and Reasoning Efficiency in LRMs
Efficient reasoning methods that skip or compress a model's thinking increase behavioral inconsistency in large reasoning models, according to a new benchmark called ICBENCH.
-
Steering LLM Thinking with Budget Guidance
A test-time budget guidance module softly biases token generation to keep LLM reasoning traces within a target token budget, improving accuracy under tight budgets and cutting tokens versus hard cutoffs.
-
NeSyFS: A Neuro-symbolic Fast-Slow Thinking Framework for LLM Agent under Partial Observability
NeSyFS combines a knowledge-graph belief state, a reflection step, and TSMC-style particle planning to improve LLM agent success rates on ALFWorld, WebShop, and ScienceWorld.
-
EvoThink: Evolving Thinking in Large Reasoning Models via Self-Pruning and Aha-Moment Preference Optimization
EvoThink reduces LRM overthinking by pruning redundant atomic reasoning steps (SPT) and training on diversity-selected wrong-to-right mutation data (AMPO), cutting tokens and improving math/code accuracy.
-
ThinkReset: Learnable Intermediate Interface Construction for Bounded-Context Long-Horizon Reasoning
Training LLMs to write reusable intermediate states before the context window fills, and optimizing directly for success after reset, improves long-horizon reasoning accuracy under fixed context windows.
-
Can Aha Moments Be Fake? Towards Quantifying Decorative and True Thinking in Chain-of-Thought
LLMs interleave true causal reasoning steps with decorative ones in CoT, with only ~2.3% of steps having high causal impact on AIME for Qwen-2.5, and a steering direction can force internal use of specific steps.
-
Are Large Reasoning Models Interruptible?
Interrupting large reasoning models or changing the problem mid-thought sharply degrades accuracy, with up to 60% drops and distinct failure modes: reasoning leakage, panic, and self-doubt.
-
AdaThink-Med: Optimizing Inference-Time Compute for Medical Reasoning via Uncertainty Quantification
An entropy-guided reward that shortens answers to easy, confident questions and extends reasoning on hard ones reduces medical QA tokens by 4.7x to 6.4x with minimal accuracy loss.
-
Scaling Up, Speeding Up: A Benchmark of Speculative Decoding for Efficient LLM Test-Time Scaling
N-gram based speculative decoding methods, especially SAM and hybrid SAM[EAGLE-3], achieve strong speedups in LLM test-time scaling by exploiting repetitive reasoning patterns.
-
Hierarchical Budget Policy Optimization for Adaptive Reasoning
Training reasoning models with hierarchical token budgets and budget-aware rewards produces up to 60.6% token reduction with no accuracy loss, and benchmark-level evidence of length adaptation.
-
NaturalThoughts: Selecting and Distilling Reasoning Traces for General Reasoning Tasks
Selecting difficult, strategy-diverse reasoning traces from DeepSeek-R1 on NaturalReasoning questions improves supervised distillation of reasoning into 7B-70B student models.
-
IMPACT: Inflectional Morphology Probes Across Complex Typologies
LLMs are systematically worse at judging ungrammatical inflectional morphology than at recognizing grammatical examples, and chain-of-thought prompting does not reliably fix this.
-
Do Thinking Tokens Help or Trap? Towards More Efficient Large Reasoning Model
Suppressing 'thinking tokens' in a 1.5B reasoning model preserves accuracy while cutting tokens, and the proposed DuP-PO RL method improves both accuracy and efficiency over GRPO.
-
AlphaOne: Reasoning Models Thinking Slow and Fast at Test Time
A stochastic scheduling framework that modulates slow-to-fast reasoning in large reasoning models at test time, improving accuracy while reducing token usage.
-
PixelThink: Towards Efficient Chain-of-Pixel Reasoning
A soft token-budget penalty driven by task difficulty and model confidence roughly halves the reasoning tokens used by segmentation MLLMs while slightly improving mask accuracy on ReasonSeg-derived benchmarks.
-
ChatVLA-2: Vision-Language-Action Model with Open-World Embodied Reasoning from Pretrained Knowledge
ChatVLA-2 uses dynamic mixture-of-experts and a two-stage training recipe to let a vision-language-action model retain pretrained reasoning while following robot instructions.
-
Reasoning LLMs are Wandering Solution Explorers
Six current reasoning LLMs, including commercial systems, exhibit structured-search failures on verifiable computation tasks and degrade as the solution space grows.
-
MMATH: A Multilingual Benchmark for Mathematical Reasoning
A new multilingual math benchmark shows that reasoning models often respond in the wrong language, and English-reasoning training improves both accuracy and language consistency.
-
Thought calibration: Efficient and confident test-time scaling
A calibrated early-exit rule for reasoning LLMs, based on hidden-state probes and Learn-then-Test risk control, reduces thinking tokens by up to 60% in-distribution and 20% out-of-distribution while roughly preserving...
-
Not All Tokens Are What You Need In Thinking
A method that scores each chain-of-thought token by answer-conditioned perplexity and trains models on the compressed traces preserves or improves reasoning accuracy with significantly fewer tokens.
-
Don't "Overthink" Passage Reranking: Is Reasoning Truly Necessary?
Standard pointwise rerankers beat reasoning-based pointwise rerankers under identical training, and removing the reasoning process at inference partially recovers the gap.
-
When to Continue Thinking: Adaptive Thinking Mode Switching for Efficient Reasoning
Adaptive Self-Recovery Reasoning (ASRR) combines a no-thinking prompt with an accuracy-gated length reward, reducing reasoning length by up to 32.5% with under 1.2 points of pass@1 loss on math benchmarks.
-
Structured Thoughts For Improved Reasoning And Context Pruning
Structured try/outcome SFT improves math reasoning by up to 8% over standard SFT and enables pruning ~85% of context with ~9% accuracy drop.
-
Is Your Model Thinking or Just Stagnating? PUMA: Diagnosing Reasoning Pathology via Phase-Momentum Alignment
PUMA detects when a reasoning model's entropy drop aligns with hidden-state momentum, truncates at that point, and reports improved accuracy-efficiency on 1.5B-32B reasoning models.
-
LISA: Linear-Indexed Sparse Attention for Efficient Long-Context Reasoning
LISA replaces standard attention with linear attention plus indexer-selected sparse attention, claiming O(nM) inference cost and accuracy gains, but the complexity analysis ignores the indexer's full-context scoring.
-
MixReasoning: Switching Modes to Think
A reasoning model can cut token usage by roughly a third to a half without losing accuracy if it switches to brief output on low-uncertainty steps and expands only at high-uncertainty forks.
-
CoRE: Enhancing Metacognition with Label-free Self-evaluation in LRMs
A training-free and label-free detector of cyclic hidden-state patterns triggers early exit during chain-of-thought reasoning, reducing token length while mostly preserving or improving accuracy.
-
Large Reasoning Models are not thinking straight: on the unreliability of thinking trajectories
Large reasoning models frequently ignore explicitly injected ground-truth solutions mid-trajectory and continue to overthink, often ending on incorrect answers.
-
EfficientXLang: Towards Improving Token Efficiency Through Cross-Lingual Reasoning
Reasoning in non-English languages reduces thinking tokens by 20-40% while largely preserving math accuracy, with savings persisting after translation to English.
-
Long or short CoT? Investigating Instance-level Switch of Large Reasoning Models
A fine-tuned selector can pick long or short chain-of-thought per question and per token budget, cutting reasoning-model inference cost by roughly 50 percent without sacrificing accuracy.
-
Deciphering Trajectory-Aided LLM Reasoning: An Optimization Perspective
Reasoning trajectories are formalized as pseudo-gradient descent on LLM parameters, making LLM reasoning training a MAML-style meta-learning problem.
-
When Can Large Reasoning Models Save Thinking? Mechanistic Analysis of Behavioral Divergence in Reasoning
QwQ-32B under a save-thinking prompt exhibits three modes, no thinking, explicit thinking, and implicit thinking, with no-thinking saving tokens but losing accuracy while explicit and implicit thinking preserve accuracy.
-
Prolonged Reasoning Is Not All You Need: Certainty-Based Adaptive Routing for Efficient LLM/MLLM Reasoning
CAR routes each query to either a short answer or full reasoning based on the perplexity of the model's draft answer, improving accuracy and cutting token use on VQA, KIE, and math/common sense benchmarks.
-
Two Experts Are All You Need for Steering Thinking: Reinforcing Cognitive Effort in MoE Reasoning Models Without Additional Training
Reinforcing the two experts most correlated with thinking tokens improves reasoning accuracy and efficiency in MoE large reasoning models, with gains of up to 10 points on AIME benchmarks.
-
ThinkSwitcher: When to Think Hard, When to Think Fast
ThinkSwitcher trains a lightweight module to predict when a reasoner should use short versus long chain-of-thought, reducing inference tokens by 20-30% with only a small accuracy drop.
-
From Long to Short: LLMs Excel at Trimming Own Reasoning Chains
EDIT searches across step-count prompts to find the shortest reasoning path the model consistently answers correctly.
-
Schema-R1: A reasoning training approach for schema linking in Text-to-SQL Task
Schema-R1 combines cold-start SFT on 200 CoT samples with GRPO rule-based RL, reporting table and column filter accuracy gains of 10 percentage points or more over a fine-tuned baseline on Spider-dev.
-
Wait, We Don't Need to "Wait"! Removing Thinking Tokens Improves Reasoning Efficiency
Suppressing "Wait"-like reflection tokens at decode time reduces reasoning token counts by 27-51% across five R1-style model families, with mixed accuracy effects.
-
Reinforcement Learning Meets Large Language Models: A Survey of Advancements and Applications Across the LLM Lifecycle
A survey that maps reinforcement learning methods, datasets, benchmarks, and open-source tools across the full training lifecycle of large language models, focusing on verifiable-reward reasoning.
-
Towards Concise and Adaptive Thinking in Large Reasoning Models: A Survey
A comprehensive review that categorizes methods for shortening and adaptively triggering chain-of-thought reasoning in large language models.
-
Evaluating Hierarchical Clinical Document Classification Using Reasoning-Based LLMs
Across 1,500 MIMIC-IV discharge summaries and 11 LLMs, no model exceeded 57% F1 on ICD-10 coding, with reasoning-labeled models slightly ahead of others, but the comparison is confounded by model differences.
- Optimizing Length Compression in Large Reasoning Models
Discussion (0). Sign in to comment.