← Back to all news
arXiv Machine Learning By Weihan Li, Junhao Wu, Xinlei Chen, Tianshi Zheng

Same Answer, Different Future: Matched Predictions Do Not Imply Matched Belief Updating in Language Models

Read the original on arXiv Machine Learning →

The paper investigates whether language models that produce the same probabilistic predictions after different types of evidence (textual or activation interventions) actually share the same underlying belief states. Using a Bernoulli inference task and an analytic Bayesian reference, the authors compare four textual routes and two activation-level interventions across three open-weight models, measuring both next-outcome continuations and explicit probabilistic estimates. They find that the format of evidence influences which readouts reflect its content, and that the effective evidence weight—how many new observations a belief is worth—varies widely across routes, with some elicitation questions altering inferred weights by orders of magnitude. In a one-shot steering scenario, an intervention that matches a natural record’s immediate prediction loses most of its effect when new evidence is introduced, demonstrating that agreement in current predictions does not guarantee equivalent belief updating.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.

Related stories

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
llms
More like this →
arXiv Computation and Language
3d ago

Multi-Estimate Elicitation Does Not Uniformly Debias Language Model Judgments

The study investigates whether prompting language models to provide five weighted estimates in a single response can reduce bias in their judgments. Across various bias tasks—anchoring, hindsight, base‑rate, conjunction, and status‑quo/default—the authors compare direct numerical answers with a multi‑estimate format, testing different source framings, random cues, and probability‑consistency constraints on three models (gpt‑6.1‑sol, claude‑sonnet‑5.5, and Qwen3‑8B). Results show that while direct elicitation reduces irrelevant‑anchor indices and status‑quo effects in two configurations, multi‑estimate elicitation does not consistently debias or attenuate responses, and the observed effects are limited by noise and lack of equivalence.

By An-Sheng Jhang
llms
More like this →
arXiv AI
Jul 1

BayesBench: Evaluating LLM Belief Trajectories Under Multi-Turn Evidence Accumulation

arXiv:2606. 30850v1 Announce Type: new Abstract: Large language models (LLMs) are typically deployed in multi-turn conversations, where each turn provides new evidence that should reduce epistemic uncertainty about their environment.

By Ankur Samanta, Akshayaa Magesh, Tal Lancewicki, Ayush Jain, Youliang Yu, Paul Sajda, Kaveh Hassani, Aditya Modi, Daniel R. Jiang, Yonathan Efroni
llms
More like this →
arXiv Machine Learning
Oct 23

Interpreting Model Reasoning through Behavioral Counterfactuals

The paper introduces a method for analyzing how interventions in a prompt affect a language model’s reasoning by using behavioral counterfactuals. It trains a module to edit the residual stream so that the next-token distributions for a fixed response match between a source and a counterfactual prompt, outperforming standard activation reconstruction. The resulting attribution maps and visualizations help identify causally important locations for counterfactual recovery and guide hypothesis testing with activation patching, as demonstrated in a case study on Gemma-4-E4B-it and DeepSeek-R1-0528-Qwen3-8B.

By Shawn Im, James Oldfield, Adel Bibi, Sharon Li
More like this →