REVIEW 18 cited by
When Thinking Fails: The Pitfalls of Reasoning for Instruction-Following in LLMs
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
Reasoning-enhanced large language models (RLLMs), whether explicitly trained for reasoning or prompted via chain-of-thought (CoT), have achieved state-of-the-art performance on many complex reasoning tasks. However, we uncover a surprising and previously overlooked phenomenon: explicit CoT reasoning can significantly degrade instruction-following accuracy. Evaluating 15 models on two benchmarks: IFEval (with simple, rule-verifiable constraints) and ComplexBench (with complex, compositional constraints), we consistently observe performance drops when CoT prompting is applied. Through large-scale case studies and an attention-based analysis, we identify common patterns where reasoning either helps (e.g., with formatting or lexical precision) or hurts (e.g., by neglecting simple constraints or introducing unnecessary content). We propose a metric, constraint attention, to quantify model focus during generation and show that CoT reasoning often diverts attention away from instruction-relevant tokens. To mitigate these effects, we introduce and evaluate four strategies: in-context learning, self-reflection, self-selective reasoning, and classifier-selective reasoning. Our results demonstrate that selective reasoning strategies, particularly classifier-selective reasoning, can substantially recover lost performance. To our knowledge, this is the first work to systematically expose reasoning-induced failures in instruction-following and offer practical mitigation strategies.
Forward citations
Cited by 18 Pith papers
-
Observable Patterns Are Not Explanations: A Causal-Geometric Analysis of Latent Reasoning Models
Evaluation of two latent reasoning models against controls shows observable latent patterns appear without the proposed mechanisms, have graded causal effects on behavior, and concentrate in structured low-rank direct...
-
More Is Not Always Better: Cross-Component Interference in LLM Agent Scaffolding
Full factorial testing of five LLM agent components reveals that the complete 'All-In' combination is consistently outperformed by smaller subsets due to cross-component interference, with optimal subsets being task- ...
-
DataDignity: Training Data Attribution for Large Language Models
ScoringModel raises mean Recall@10 to 52.2 on the FakeWiki provenance benchmark from 35.0 for the best baseline, winning 41 of 45 model-by-condition comparisons and gaining 15.7 points on jailbreak-style queries.
-
When Reasoning Hurts Legal Drafting: The Verbalization Bottleneck in Patent Claim Generation
Implicit CoT consistently outperforms explicit CoT on patent claim generation because verbalized reasoning abstracts details, disrupts drafting patterns, and cascades early errors.
-
When Built-in Thinking Helps and Hurts: Constraint-Level Error Shifts in Instruction Following
Thinking mode in Qwen3 models improves class-level performance on planning constraints but worsens precision constraints in IFEval, with 10-20% prompt-level flips and directional consistency in Hunyuan models.
-
CLORE: Content-Level Optimization for Reasoning Efficiency
CLORE augments correct on-policy rollouts by deleting repetitive and irrelevant segments then optimizes with auxiliary DPO to improve accuracy-efficiency trade-off on math benchmarks.
-
Chain of Risk: Safety Failures in Large Reasoning Models and Mitigation via Adaptive Multi-Principle Steering
Reasoning traces in large reasoning models expose safety failures missed by final-answer checks, and adaptive multi-principle steering reduces unsafe content in both traces and answers while preserving task performance.
-
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.
-
Retrieval-of-Thought: Efficient Reasoning via Reusing Thoughts
Retrieval-of-Thought organizes prior reasoning into a thought graph for retrieval and reward-guided recombination, reducing output tokens by up to 40% and latency by 82% while preserving accuracy on reasoning benchmarks.
-
ProactiveEval: A Unified Evaluation Framework for Proactive Dialogue Agents
A unified evaluation framework for proactive dialogue agents, built with 328 synthetic environments across six domains, shows that thinking modes improve target planning but not dialogue guidance in a 22-model comparison.
-
The Illusion of Thinking: Understanding the Strengths and Limitations of Reasoning Models via the Lens of Problem Complexity
LRMs exhibit complete accuracy collapse beyond certain puzzle complexities, with reasoning effort rising then declining, outperforming standard LLMs only on medium-complexity tasks.
-
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.
-
M3CoTBench: Benchmark Chain-of-Thought of MLLMs in Medical Image Understanding
A medical-image benchmark that scores the step-by-step reasoning chains of multimodal LLMs shows current models explain poorly and chain-of-thought prompting frequently reduces diagnostic accuracy.
-
Position: The Hidden Costs and Measurement Gaps of Reinforcement Learning with Verifiable Rewards
The paper identifies confounds in RLVR evaluations that inflate apparent gains and proposes a minimum standard for budget-matched, contamination-aware assessment with calibration tracking.
-
"GenAI Defaults to Bias!" Gamify AI Literacy Through Reflections on Prompts
Playing ImaginAItion, a prompt-minimization party game, helped 30 adults recognize GenAI default biases and adjust their prompting strategies, according to pre-post survey coding.
-
AutoTIR: Autonomous Tools Integrated Reasoning via Reinforcement Learning
AutoTIR applies GRPO with a hand-designed action reward so a 7B instruct model learns to mix search and code tools, beating tool-using baselines on ten benchmarks.
-
Prompt Governance? On Governing Technologies Governed by Natural Language
Literature on system prompts for AI shows fragmented and contradictory claims that complicate policy efforts to use them as reliable governance mechanisms.
-
A Survey of Reinforcement Learning for Large Reasoning Models
A survey compiling RL methods, challenges, data resources, and applications for enhancing reasoning in large language models and large reasoning models since DeepSeek-R1.
Discussion (0). Sign in to comment.