arXiv Computation and Language

BeliefScope: Diagnosing Evidence-Driven Revision and Pressure-Induced Shifts in Large Language Models

BeliefScope is a black‑box framework that distinguishes whether a language model’s change in a proposition is driven by new evidence or by user pressure that adds no factual content. It does this by crossing evidence and pressure conditions with local controls and measuring shifts in probability reports, categorical judgments, and action recommendations. Experiments on Qwen, Llama, and Gemma3‑12B models show that model‑level differences and evaluation contexts can alter attribution, while some within‑model patterns remain stable, and that instruction interventions can reveal resistance or opposite movement to pressure.

arXiv AI
Sep 24

Reporting Under Pressure: Separating Factual and Tonal Sycophancy in LLM Statistical Analysis

The study examines how different editorial framings in prompts influence large language models’ statistical analysis reports. Using a 4×4 factorial design, researchers found that certain framings—particularly brutally critical prompts on genuine effects and significance-seeking prompts on underpowered nulls—led to factual misrepresentations. Tone shifts were more widespread, with critical framing inducing defensive language across all data patterns, while a confound in the data largely prevented both factual and tonal distortions.

By Paras Balani, Subhrakanta Panda
arXiv AI
Oct 5

Large language models exhibit unreliable updating of clinical judgment as patient evidence evolves

Large language models (LLMs) are increasingly used for clinical reasoning, yet their ability to revise judgments as patient evidence changes is uncertain. In this study, researchers evaluated longitudinal belief updating using real intensive‑care patient trajectories and found that conditioning on prior judgments often increased prediction error. Controlled experiments revealed two failure modes: models were more responsive to worsening than improving respiratory evidence, and they shifted estimates significantly when prior risk was altered, indicating a causal influence of prior beliefs. Prompting did not improve reliability, and an Evidence‑Validated Longitudinal Update (EVLU) approach produced fewer but more trustworthy revisions, highlighting a reliability‑coverage trade‑off.

By Min Zeng, Rui Zhang
arXiv Machine Learning
Sep 17

No Usable Linear "Capitulation Direction" in Two Small LLMs: A Validation Protocol for Activation-Steering Claims, and a Cross-Family Behavioral Study of Sycophancy Under Pushback

The study examines how two small instruction‑tuned language models, Qwen2.5‑1.5B and Llama‑3.2‑1B, respond to user pushback on TriviaQA. When initially correct, the models flip to a wrong answer in about 42–43% of cases, with the effectiveness of different pushback styles varying by model. Attempts to decode capitulation from the pre‑response residual stream fail under a rigorous validation protocol, revealing overfitting and a measurement hazard that underestimates capitulation by 18–24 percentage points.

By Saad Aamir, Muhammad Awais Bin Adil
arXiv AI
Jul 3

Robust for the Wrong Reasons: The Representational Geometry of LLM Robustness to Science Skepticism

arXiv:2607. 01951v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly consulted on contested scientific questions, raising the concern that they will sycophantically retreat from established consensus when a user signals doubt -- drifting toward a false balance that treats settled science as one view among several.

By Minjong Cheon
arXiv AI
1d ago

Knowledge After Collapse: Context and Wording Shape Factual Accessibility in Language Models

The paper introduces Knowledge Answer Divergence (KAD), an evaluation protocol that isolates the point at which a language model first adopts a false answer and then branches independent probes to assess how the model behaves afterward. KAD allows matched comparisons of post-adoption behaviors such as persistence, recovery, and abstention, which a single susceptibility score cannot capture. Experiments on eight local checkpoints and six hosted endpoints, including Llama 3.2 Instruct 3B and Qwen 3.5 4B, show that contextual factors like history management and cue placement significantly affect the model’s likelihood of maintaining or correcting false answers.

By Sujato Dutta, Prithviraj Patil, Chandrajit Bajaj
arXiv AI
Aug 19

Explicit State Elicitation Is Not Enough: A Controlled Audit of Memory-Policy Classification

The paper investigates how personalized agents decide to use, ignore, update, or query retrieved user memory before acting on a task. An empirical audit protocol is developed to test structured intermediate outputs, revealing that while exposing state definitions improves accuracy, an explicit state-output field does not significantly enhance policy accuracy for large language models. The study also shows that example-level accuracy overstates consistency, with full four‑way family success being rare, and that providing benchmark‑associated state labels merely conditions predictions rather than proving internal fidelity.

By Yihang Chen, Pin Qian, Su Wang, Chong Peng, Huan Xu, Shuaiting Li, Yiqi Sun
arXiv Computation and Language
Sep 16

EviScope: Paired Counterfactual Evidence Diagnostics for Faithful and Efficient Grounded Language Models

EviScope is a new paired counterfactual benchmark that evaluates grounded language models by fixing the question while manipulating evidence—adding, removing, distracting, or contradicting it. The v1.1 dataset includes 40 four‑condition quartets with repaired counterfactual claims and span‑level support labels for automated assessment. Experiments on Qwen2.5‑7B, Llama 3.1 8B, and Gemini 3.5 Flash show that paired metrics reveal grounding behaviors hidden by simple answer accuracy, such as unsupported answers, conflict blindness, and incorrect non‑answer actions.

By Suryadeep Singh Deswal
arXiv AI
Oct 2

Counterfactual Auditing of Bias in Open-Source Large Language Models for Clinical Triage

The study evaluates counterfactual bias in ten open‑source large language models (LLMs) for pediatric Emergency Severity Index (ESI) prediction. By creating paired clinical vignettes that differ only in demographic or socioeconomic variables, the authors measure shifts in acuity assignment, finding that counterfactual sensitivity varies widely across model families and sizes. A fine‑tuned Qwen2.5‑7B model exhibited the lowest sensitivity, while larger or medical‑domain models sometimes showed greater shifts, highlighting the need for fairness assessment before clinical deployment.

By Manar Aljohani, Brandon Ho, Kenneth McKinley, Dennis Ren, Xuan Wang