More Thinking, More Bias: Length-Driven Position Bias in Reasoning Models

• URL: http://arxiv.org/abs/2605.06672v1
• Date: Tue, 21 Apr 2026 04:14:13 GMT
• Title: More Thinking, More Bias: Length-Driven Position Bias in Reasoning Models
• Abstract summary: Chain-of-thought (CoT) reasoning and reasoning-tuned models are commonly assumed to reduce shallow biases by thinking carefully.<n>We test this on position bias in multiple-choice QA and find a different story: within any reasoning-capable model, per-question position bias scales with the length of the reasoning trajectory.
• Score: 5.705685936981751
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Chain-of-thought (CoT) reasoning and reasoning-tuned models such as DeepSeek-R1 are commonly assumed to reduce shallow heuristic biases by thinking carefully. We test this on position bias in multiple-choice QA and find a different story: within any reasoning-capable model, per-question position bias scales with the length of the reasoning trajectory. Across thirteen reasoning-mode configurations (two R1-distilled 7-8B models, two base models prompted with CoT, and DeepSeek-R1 at 671B) on MMLU, ARC-Challenge, and GPQA, twelve show a positive partial correlation between trajectory length and Position Bias Score (PBS) after controlling for accuracy, ranging from 0.11 to 0.41 (all p < 0.05). All twelve open-weight reasoning-mode configurations show monotonically increasing PBS across length quartiles. A truncation intervention provides causal evidence: continuations resumed from later points in the trajectory are increasingly likely to shift toward position-preferred options (16% to 32% for R1-Qwen-7B across absolute-position buckets). At 671B, aggregate PBS collapses to 0.019, but the length effect still manifests in the longest quartile (PBS = 0.071), suggesting that accuracy gates the expression of length-driven bias rather than eliminating the underlying mechanism. We additionally find that direct-answer position bias is a distinct phenomenon with a different footprint (strong in Llama-Instruct-direct, weak in Qwen-Instruct-direct, and uncorrelated with trajectory length): CoT reasoning replaces this baseline bias with length-accumulated bias. Our results argue that reasoning-capable models should not be treated as order-robust by default in MCQ evaluation pipelines, and offer a diagnostic toolkit (PBS, commitment change point, effective switching, truncation probes) for auditing position bias in reasoning models.

Related papers

• How Does Alignment Tuning Shape Representations of Sycophancy and Related Cue-Induced Biases in LLMs? [53.58941416780391]
  We study where this susceptibility, spanning sycophancy and related cue-induced biases, lives inside the model.<n>We extract a per-bias direction from hidden states and triangulate it through three measures: probing, leave-one-dataset-out transfer, and causal intervention.
  arXiv  Detail & Related papers  (2026-07-20T16:10:06Z)
• When Built-in Thinking Helps and Hurts: Constraint-Level Error Shifts in Instruction Following [0.0]
  Large reasoning models (LRMs) often improve math and coding performance, but their effect on instruction following is unclear.<n>We study IFEval with Qwen3 models (1.7B-32B) using same-weights Thinking ON/OFF controls.
  arXiv  Detail & Related papers  (2026-06-08T15:45:04Z)
• On the Role of Reasoning Patterns in the Generalization Discrepancy of Long Chain-of-Thought Supervised Fine-Tuning [63.41902113656453]
  Supervised Fine-Tuning (SFT) on long Chain-of-Thought (CoT) trajectories has become a pivotal phase in building large reasoning models.<n>We conduct a comparative study using two sources of verified CoT trajectories generated by two competing models.<n>We find that SFT on textttDeepSeek-R1-0528 data achieves remarkably lower training loss, yet exhibits significantly worse generalization performance.
  arXiv  Detail & Related papers  (2026-04-02T07:00:54Z)
• Directional Reasoning Trajectory Change (DRTC): Identifying Critical Trace Segments in Reasoning Models [0.0]
  Directional Reasoning Trajectory Change (DRTC) is a process-causal framework for interpreting long-form reasoning from a single on-policy rollout.<n>DRTC detects pivot decision points using uncertainty and distribution-shift signals.<n>It measures whether each intervention redirects the direction of the model's log-probability trajectory.
  arXiv  Detail & Related papers  (2026-02-17T03:38:16Z)
• Decoupling the Effect of Chain-of-Thought Reasoning: A Human Label Variation Perspective [60.45433515408158]
  We show that long Chain-of-Thought (CoT) serves as a decisive decision-maker for the top option but fails to function as a granular distribution calibrator for ambiguous tasks.<n>We observe a distinct "decoupled mechanism": while CoT improves distributional alignment, final accuracy is dictated by CoT content.
  arXiv  Detail & Related papers  (2026-01-06T16:26:40Z)
• The Procrustean Bed of Time Series: The Optimization Bias of Point-wise Loss [53.542743390809356]
  This paper aims to provide a first-principles analysis of the Expectation of Optimization Bias (EOB)<n>Our analysis reveals a fundamental paradigm paradox: the more deterministic and structured the time series, the more severe the bias by point-wise loss function.<n>We present a concrete solution that simultaneously achieves both principles via DFT or DWT.
  arXiv  Detail & Related papers  (2025-12-21T06:08:22Z)
• DTS: Enhancing Large Reasoning Models via Decoding Tree Sketching [54.98126916293868]
  Large Reasoning Models (LRMs) produce excessively long chain-of-thought traces that degrade accuracy.<n>We propose a model-agnostic decoding framework that sketches the reasoning space by branching at high-entropy tokens and applies early stopping to select the shortest completed reasoning path.<n>This approach approximates the optimal solution that enhances both efficiency and accuracy, without requiring additional training or supervision.
  arXiv  Detail & Related papers  (2025-11-01T17:41:28Z)
• SEAL: Steerable Reasoning Calibration of Large Language Models for Free [58.931194824519935]
  Large Language Models (LLMs) have demonstrated compelling capabilities for complex reasoning tasks via the extended chain-of-thought (CoT) reasoning mechanism.<n>Recent studies reveal substantial redundancy in the CoT reasoning traces, which negatively impacts model performance.<n>We introduce SEAL, a training-free approach that seamlessly calibrates the CoT process, improving accuracy while demonstrating significant efficiency gains.
  arXiv  Detail & Related papers  (2025-04-07T02:42:07Z)
• Does Reasoning Introduce Bias? A Study of Social Bias Evaluation and Mitigation in LLM Reasoning [21.921684911834447]
  We present the first systematic evaluation of social bias within large language models (LLMs)<n>We analyze both prediction accuracy and reasoning bias across a broad spectrum of models, including instruction-tuned and CoT-augmented variants of DeepSeek-R1 and ChatGPT.<n>We propose Answer Distribution as Bias Proxy (ADBP), a lightweight mitigation method that detects bias by tracking how model predictions change across incremental reasoning steps.
  arXiv  Detail & Related papers  (2025-02-21T10:16:07Z)
• Bias-Augmented Consistency Training Reduces Biased Reasoning in Chain-of-Thought [33.32335629744919]
  Chain-of-thought prompting (CoT) has the potential to improve the explainability of language model reasoning.<n>CoT can also systematically misrepresent the factors influencing models' behavior.<n>We first create a new dataset of 9 different biases that affect GPT-3.5-Turbo and Llama-8b models.
  arXiv  Detail & Related papers  (2024-03-08T18:41:42Z)

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.