OptimismBench: Forecasting Bias and the Alignment Effect in Language Model Judgment
- URL: http://arxiv.org/abs/2607.26981v1
- Date: Wed, 29 Jul 2026 14:38:55 GMT
- Title: OptimismBench: Forecasting Bias and the Alignment Effect in Language Model Judgment
- Abstract summary: We introduce OptimismBench, which detects directional bias with inverted pairs.<n>We show post-training sets the sign of the bias, with opposite shifts in different families.<n>We release 3,870 items across 10 languages for per-model directional-bias auditing.
- Score: 0.9821874476902969
- License: http://creativecommons.org/licenses/by/4.0/
- Abstract: Large language models are increasingly used as decision aids whose probability judgments shape downstream choices. Whether those judgments carry a systematic directional tilt has been hard to detect: calibration metrics aggregate unsigned errors, and naturalistic uncertainty offers no ground-truth probability. When an LLM rates a startup's success at 70% but its failure at 15%, the missing 15 points expose a distortion no aggregate score flags. We introduce OptimismBench, which detects directional bias with inverted pairs: each scenario elicits both P(success) and P(failure), and asymmetry between the two framings yields a signed bias score without ground truth. Across 16 models from 8 providers, fourteen are optimistic; pessimism appears only in Anthropic's frontier tier. Eleven matched base-versus-chat pairs across four families show post-training sets the sign of the bias, with opposite shifts in different families. The pattern survives prompt, temperature, perspective, and self-debiasing ablations. A seventeen-model six-language comparison further shows model identity dominates language, with inter-model variance at 4.7x inter-language variance. We release 3,870 items across 10 languages for per-model directional-bias auditing. When alignment makes a model more helpful, it also tilts its probabilities; downstream pipelines inherit the tilt by default.
Related papers
- Marginal Fidelity Does Not Establish User Simulation in Demographic Synthetic Survey Panels: Response Contracts, Support Collapse and Conditioning Failure [51.736723807086385]
Population-marginal agreement is evidence about an elicitation contract and an estimand obtainable without simulated respondents, not evidence of individual simulation.<n>On nine aligned model-battery pairs, a no-persona population-prevalence query averages 6.27 MAE versus 12.39 for committed panels and wins all nine comparisons.
arXiv Detail & Related papers (2026-09-07T10:22:22Z) - Interpretable Cross-Lingual Alignment in Small Language Models: Probing Cultural and Pragmatic Reasoning in Japanese-English Bilingual LLMs [0.0]
I introduce J-PragEval-v0, a minimal-pair benchmark isolating four phenomena from surface fluency.<n>I combine it with linear probes and teacher-forced log-probability evaluation to ask where inside TinySwallow-1.5B the corresponding contrasts live.
arXiv Detail & Related papers (2026-08-14T21:04:21Z) - Actions Speak Louder than Words: Measuring Cross-Lingual Policy Retention in Tool-Using Agents [22.30468215041597]
A tool-using agent is given the same task in a different language, does it still take the same steps?<n>We make the action policy the measured object across 8 models, 6 parallel benchmarks and 41 languages.
arXiv Detail & Related papers (2026-08-11T16:18:34Z) - 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) - Decodable Is Not Grounded: A Vision-Ablation Arbiter for VLM Spatial Reasoning [48.438568700670835]
The standard way to read latent knowledge out of a model can systematically overstate what a vision-language model grounds in the image.<n>We show this on spatial reasoning, where the error is invisible to both probing and steering yet exposed by a one-line causal control.
arXiv Detail & Related papers (2026-06-30T07:33:18Z) - 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) - How reliable are LLMs when it comes to playing dice? [0.0]
We investigate the reasoning capabilities of large language models through a controlled benchmarking study on discrete probability problems.<n>Models achieve an average accuracy of 0.96 on standard problems but only 0.59 on counterintuitive ones.<n> Embedding misleading suggestions in the prompt reduces performance by up to 34%, with no model proving immune.
arXiv Detail & Related papers (2026-06-05T17:59:42Z) - Judging the Judges: A Systematic Evaluation of Bias Mitigation Strategies in LLM-as-a-Judge Pipelines [0.0]
We compare nine debiasing strategies across five judge models from four provider families.<n>Style bias is the dominant bias (0.76-0.92 across all models), far exceeding position bias (= 0.04)<n>The combined budget strategy significantly improves Claude Sonnet 4 by +11.2 pp (p 0.0001), with directionally positive trends for other models.
arXiv Detail & Related papers (2026-04-25T07:18:30Z) - The Model Says Walk: How Surface Heuristics Override Implicit Constraints in LLM Reasoning [9.898274894485107]
Large language models systematically fail when a salient surface cue conflicts with an unstated feasibility constraint.<n>We study this through a diagnose-measure-bridge-treat framework.
arXiv Detail & Related papers (2026-03-30T21:36:09Z) - CLUE: Non-parametric Verification from Experience via Hidden-State Clustering [64.50919789875233]
We show that correctness of a solution is encoded as a geometrically separable signature within the trajectory of hidden activations.<n>ClUE consistently outperforms LLM-as-a-judge baselines and matches or exceeds modern confidence-based methods in reranking candidates.
arXiv Detail & Related papers (2025-10-02T02:14:33Z) - Judging with Confidence: Calibrating Autoraters to Preference Distributions [56.17041629492863]
We argue that a reliable autorater must learn to model the full distribution of preferences defined by a target population.<n>We present two learning methods tailored to different data conditions.<n>Our results show that finetuning autoraters with a distribution-matching objective leads to verbalized probability predictions that are better aligned with the target preference distribution.
arXiv Detail & Related papers (2025-09-30T20:36:41Z) - Eliminating Position Bias of Language Models: A Mechanistic Approach [119.34143323054143]
Position bias has proven to be a prevalent issue of modern language models (LMs)<n>Our mechanistic analysis attributes the position bias to two components employed in nearly all state-of-the-art LMs: causal attention and relative positional encodings.<n>By eliminating position bias, models achieve better performance and reliability in downstream tasks, including LM-as-a-judge, retrieval-augmented QA, molecule generation, and math reasoning.
arXiv Detail & Related papers (2024-07-01T09:06:57Z) - Holistic Evaluation of Language Models [183.94891340168175]
Language models (LMs) are becoming the foundation for almost all major language technologies, but their capabilities, limitations, and risks are not well understood.
We present Holistic Evaluation of Language Models (HELM) to improve the transparency of language models.
arXiv Detail & Related papers (2022-11-16T18:51:34Z)
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.