ECHO is a clipped policy-gradient method that uses posterior-sensitive rewards to give turn-level epistemic credit in multi-turn information-seeking tasks, outperforming trajectory-level GRPO on a new Clue Selector Game benchmark.
BED-LLM: Intelligent Information Gathering with LLMs and Bayesian Experimental Design
10 Pith papers cite this work. Polarity classification is still indexing.
10
Pith papers citing it
abstract
We propose a general-purpose approach for improving the ability of large language models (LLMs) to intelligently and adaptively gather information from a user or other external source using the framework of sequential Bayesian experimental design (BED). This enables LLMs to act as effective multi-turn conversational agents and interactively interface with external environments. Our approach, which we call BED-LLM (Bayesian experimental design with large language models), is based on iteratively choosing questions or queries that maximize the expected information gain (EIG) with respect to a variable of interest given the responses gathered previously. We show how this EIG can be formulated (and then estimated) in a principled way using a probabilistic model derived from the LLM's predictive distributions and provide detailed insights into key decisions in its construction and updating procedure. We find that BED-LLM achieves substantial gains in performance across a wide range of tests based on the 20 Questions game and using the LLM to actively infer user preferences, compared to purely prompting-based design generation and other adaptive design strategies.
citation-role summary
background 1
citation-polarity summary
years
2026 10roles
background 1polarities
background 1representative citing papers
This paper introduces a systems-level conceptual framing and a three-level taxonomy (intra-model, system-level, socio-technical) for uncertainty propagation in compound LLM applications, along with engineering insights and open challenges.
CA-BED uses Bayesian experimental design and simulated conversation trees with LLM likelihoods to optimize multi-turn question selection, reporting 21.8% higher success rates than direct prompting on entity-deduction benchmarks.
TPA, a proactive multi-agent dialogue system, achieves 82.1% SLD trait coverage in simulated ADOS-2 assessments, outperforming real clinician dialogues by 16.6% and other AI baselines.
LLMs show inconsistent belief updates from evidence, with learned heuristics sometimes beating exact Bayesian computation due to misspecified world models.
VLMs suffer from a perceptual bandwidth bottleneck; the paper formalizes active visual reasoning as sequential Bayesian optimal experimental design, derives a coverage-resolution proxy objective, and introduces the training-free FOVEA method that yields gains on high-resolution benchmarks.
Separating language from Bayesian reasoning lets cheap LLM sensors beat larger standalone LLM doctors on conversational diagnosis with controllable abstention and lower cost.
CovQValue achieves 51-77% higher branch coverage than greedy baselines on TestGenEval Lite by using coverage feedback and LLM-estimated Q-values to select informative test plans.
The paper introduces an Information Gain Reward to train clarification behavior in LLM agents, reporting a 3.7% success rate gain over no-clarification baselines in τ-Bench evaluations across five models with minimal added steps.
Proposes cost-aware question selection for ambiguous object navigation via information-gain analysis on corpora, a cost-penalizing benchmark, and a zero-shot MLLM agent.
citing papers explorer
-
ECHO: Learning Epistemically Adaptive Language Agents with Turn-Level Credit
ECHO is a clipped policy-gradient method that uses posterior-sensitive rewards to give turn-level epistemic credit in multi-turn information-seeking tasks, outperforming trajectory-level GRPO on a new Clue Selector Game benchmark.
-
Uncertainty Propagation in LLM-Based Systems
This paper introduces a systems-level conceptual framing and a three-level taxonomy (intra-model, system-level, socio-technical) for uncertainty propagation in compound LLM applications, along with engineering insights and open challenges.
-
CA-BED: Conversation-Aware Bayesian Experimental Design
CA-BED uses Bayesian experimental design and simulated conversation trees with LLM likelihoods to optimize multi-turn question selection, reporting 21.8% higher success rates than direct prompting on entity-deduction benchmarks.
-
A Proactive Multi-Agent Dialogue Framework for Assessing Social Language Disorder Traits in Autism
TPA, a proactive multi-agent dialogue system, achieves 82.1% SLD trait coverage in simulated ADOS-2 assessments, outperforming real clinician dialogues by 16.6% and other AI baselines.
-
LLMs are not (consistently) Bayesian: Quantifying internal (in)consistencies of LLMs' probabilistic beliefs
LLMs show inconsistent belief updates from evidence, with learned heuristics sometimes beating exact Bayesian computation due to misspecified world models.
-
The Perceptual Bandwidth Bottleneck in Vision-Language Models: Active Visual Reasoning via Sequential Experimental Design
VLMs suffer from a perceptual bandwidth bottleneck; the paper formalizes active visual reasoning as sequential Bayesian optimal experimental design, derives a coverage-resolution proxy objective, and introduces the training-free FOVEA method that yields gains on high-resolution benchmarks.
-
MoBayes: A Modular Bayesian Framework for Separating Reasoning from Language in Conversational Clinical Decision Support
Separating language from Bayesian reasoning lets cheap LLM sensors beat larger standalone LLM doctors on conversational diagnosis with controllable abstention and lower cost.
-
Planning to Explore: Curiosity-Driven Planning for LLM Test Generation
CovQValue achieves 51-77% higher branch coverage than greedy baselines on TestGenEval Lite by using coverage feedback and LLM-estimated Q-values to select informative test plans.
-
Uncertainty-Aware Clarification in LLM Agents with Information Gain
The paper introduces an Information Gain Reward to train clarification behavior in LLM agents, reporting a 3.7% success rate gain over no-clarification baselines in τ-Bench evaluations across five models with minimal added steps.
-
Ask When It Pays: Cost-Aware Open-Ended Interaction for Instance Goal Navigation
Proposes cost-aware question selection for ambiguous object navigation via information-gain analysis on corpora, a cost-penalizing benchmark, and a zero-shot MLLM agent.