mlsanityData to 7 Oct 2026

DualStake: Dual-Path Confidence Calibration in Deep Research Agents

arXiv 2609.00935 ↗First posted 1 Sep 2026cs.CL cs.AI cs.LGcited by 0

On the map: LLM agents and tool use → Web, search, and research agents

Abstract (arXiv)

Deep Research agents tackle knowledge-intensive tasks through multi-round retrieval and decision-oriented generation. However, these agents suffer from severe overconfidence, making their expressed confidence unreliable for user trust and downstream abstention. To address this, we augment the Deep Research pipeline with step confidence elicitation after each retrieval, building on the commonly used post-answer verbalized confidence. Interestingly, we find that Evidence Confidence (E-Conf), elicited after the final retrieval step, provides a stronger uncertainty signal than Answer Confidence (A-Conf), elicited after answer generation, and that A-Conf is largely shaped by E-Conf. Based on these findings, we propose DualStake, a dual-path calibration method that applies margin-clipped, confidence-dependent stake rewards to jointly align E-Conf and A-Conf with answer correctness while limiting extreme confidence optimization. Experiments on Qwen2.5-7B, Qwen2.5-7B-Instruct, and Qwen3-4B across 8 QA benchmarks demonstrate that DualStake consistently improves calibration without sacrificing answer accuracy. The code is available at https://github.com/FloXXXt/DualStake.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Web, search, and research agents, one of 16 directions in LLM agents and tool use. Builds on 24 of its 38 references that are in the map. First posted on arXiv 1 Sep 2026.

Builds on

24 of 38 references.

Closest ancestors · 2

Same direction, newest first

  1. 2026How to Train Your Deep Research Agent? Prompt, Reward, and Policy Optimization in Search-R123 Feb
  2. 2025DeepResearch-Slice: Bridging the Retrieval-Utilization Gap via Explicit Text Slicing16 Dec

Other references · 22

  1. 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Taskscited by 9,160
  2. 2017On Calibration of Modern Neural Networkscited by 4,860
  3. 2020Dense Passage Retrieval for Open-Domain Question Answeringcited by 3,779
  4. 2018HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answeringcited by 3,686
  5. 2017TriviaQA: A Large Scale Distantly Supervised Challenge Dataset for Reading Comprehensioncited by 2,895
  6. 2022Language Models (Mostly) Know What They Knowcited by 1,783
  7. 2025Search-R1: Training LLMs to Reason and Leverage Search Engines with Reinforcement Learningcited by 1,371
  8. 2020Constructing A Multi-hop QA Dataset for Comprehensive Evaluation of Reasoning Stepscited by 1,259
  9. 2022Text Embeddings by Weakly-Supervised Contrastive Pre-trainingcited by 1,216
  10. 2022When Not to Trust Language Models: Investigating Effectiveness of Parametric and Non-Parametric Memoriescited by 1,103
  11. 2022Measuring and Narrowing the Compositionality Gap in Language Modelscited by 1,054
  12. 2023Can LLMs Express Their Uncertainty? An Empirical Evaluation of Confidence Elicitation in LLMscited by 946
10 more
  1. 2023Just Ask for Calibration: Strategies for Eliciting Calibrated Confidence Scores from Language Models Fine-Tuned with Human Feedbackcited by 788
  2. 2022Teaching Models to Express Their Uncertainty in Wordscited by 731
  3. 2025BrowseComp: A Simple Yet Challenging Benchmark for Browsing Agentscited by 616
  4. 2023Towards Best Practices of Activation Patching in Language Models: Metrics and Methodscited by 328
  5. 2025ZeroSearch: Incentivize the Search Capability of LLMs without Searchingcited by 203
  6. 2022Calibrating Sequence likelihood Improves Conditional Language Generationcited by 121
  7. 2025Beyond Binary Rewards: Training LMs to Reason About Their Uncertaintycited by 107
  8. 2025Rewarding Doubt: A Reinforcement Learning Approach to Calibrated Confidence Expression of Large Language Modelscited by 48
  9. 2026The Confidence Dichotomy: Analyzing and Mitigating Miscalibration in Tool-Use Agentscited by 14
  10. 2026Agentic Confidence Calibrationcited by 12

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026Evaluating Confidence-Gated Retrieval with Matched Trajectory Replay
  2. 2026Towards Trustworthy Report Generation: A Deep Research Agent with Progressive Confidence Estimation and Calibration
  3. 2026From Intent to Evidence: A Categorical Approach for Structural Evaluation of Deep Research Agents
  4. 2025Refine and Align: Confidence Calibration through Multi-Agent Interaction in VQA
  5. 2025Read Your Own Mind: Reasoning Helps Surface Self-Confidence Signals in LLMs
  6. 2026ORCE: Order-Aware Alignment of Verbalized Confidence in Large Language Models
  7. 2025Don't Think Twice! Over-Reasoning Impairs Confidence Calibration
  8. 2026On the Pitfalls of Verbalized Confidence Priors for Calibrating Large Reasoning Models