Thinking Makes LLM Agents Introverted: How Mandatory Thinking Can Backfire in User-Engaged Agents

• URL: http://arxiv.org/abs/2602.07796v1
• Date: Sun, 08 Feb 2026 03:23:22 GMT
• Title: Thinking Makes LLM Agents Introverted: How Mandatory Thinking Can Backfire in User-Engaged Agents
• Authors: Jiatong Li, Changdae Oh, Hyeong Kyu Choi, Jindong Wang, Sharon Li,
• Abstract summary: Eliciting reasoning has emerged as a powerful technique for improving the performance of large language models (LLMs) on complex tasks by inducing thinking.<n>We conduct a comprehensive study on the effect of explicit thinking in user-engaged LLM agents.<n>We find that mandatory thinking often backfires on agents in user-engaged settings, causing anomalous performance degradation.
• Score: 23.785816075149484
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Eliciting reasoning has emerged as a powerful technique for improving the performance of large language models (LLMs) on complex tasks by inducing thinking. However, their effectiveness in realistic user-engaged agent scenarios remains unclear. In this paper, we conduct a comprehensive study on the effect of explicit thinking in user-engaged LLM agents. Our experiments span across seven models, three benchmarks, and two thinking instantiations, and we evaluate them through both a quantitative response taxonomy analysis and qualitative failure propagation case studies. Contrary to expectations, we find that mandatory thinking often backfires on agents in user-engaged settings, causing anomalous performance degradation across various LLMs. Our key finding reveals that thinking makes agents more ``introverted'' by shortening responses and reducing information disclosure to users, which weakens agent-user information exchange and leads to downstream task failures. Furthermore, we demonstrate that explicitly prompting for information disclosure reliably improves performance across diverse model families, suggesting that proactive transparency is a vital lever for agent optimization. Overall, our study suggests that information transparency awareness is a crucial yet underexplored perspective for the future design of reasoning agents in real-world scenarios. Our code is available at https://github.com/deeplearning-wisc/Thinking-Agent.

Related papers

• Pushing Forward Pareto Frontiers of Proactive Agents with Behavioral Agentic Optimization [61.641777037967366]
  Proactive large language model (LLM) agents aim to actively plan, query, and interact over multiple turns.<n>Agentic reinforcement learning (RL) has emerged as a promising solution for training such agents in multi-turn settings.<n>We propose BAO, an agentic RL framework that combines behavior enhancement to enrich proactive reasoning and information-gathering capabilities.
  arXiv  Detail & Related papers  (2026-02-11T20:40:43Z)
• Gaming the Judge: Unfaithful Chain-of-Thought Can Undermine Agent Evaluation [76.5533899503582]
  Large language models (LLMs) are increasingly used as judges to evaluate agent performance.<n>We show this paradigm implicitly assumes that the agent's chain-of-thought (CoT) reasoning faithfully reflects both its internal reasoning and the underlying environment state.<n>We demonstrate that manipulated reasoning alone can inflate false positive rates of state-of-the-art VLM judges by up to 90% across 800 trajectories spanning diverse web tasks.
  arXiv  Detail & Related papers  (2026-01-21T06:07:43Z)
• Are Your Agents Upward Deceivers? [73.1073084327614]
  Large Language Model (LLM)-based agents are increasingly used as autonomous subordinates that carry out tasks for users.<n>This raises the question of whether they may also engage in deception, similar to how individuals in human organizations lie to superiors to create a good image or avoid punishment.<n>We observe and define agentic upward deception, a phenomenon in which an agent facing environmental constraints conceals its failure and performs actions that were not requested without reporting.
  arXiv  Detail & Related papers  (2025-12-04T14:47:05Z)
• Cognitive Inception: Agentic Reasoning against Visual Deceptions by Injecting Skepticism [81.39177645864757]
  We propose textbfInception, a fully reasoning-based agentic reasoning framework to conduct authenticity verification by injecting skepticism.<n>To the best of our knowledge, this is the first fully reasoning-based framework against AIGC visual deceptions.
  arXiv  Detail & Related papers  (2025-11-21T05:13:30Z)
• Simulating Misinformation Vulnerabilities With Agent Personas [1.0120858915885353]
  We develop an agent-based simulation using Large Language Models to model responses to misinformation.<n>We construct agent personas spanning five professions and three mental schemas, and evaluate their reactions to news headlines.<n>Our findings show that LLM-generated agents align closely with ground-truth labels and human predictions, supporting their use as proxies for studying information responses.
  arXiv  Detail & Related papers  (2025-10-31T18:44:00Z)
• JudgeAgent: Knowledge-wise and Dynamic LLM Evaluation with Agent-as-Interviewer [19.09571232466437]
  We propose Agent-as-Interviewer, a dynamic evaluation paradigm for large language models (LLMs)<n>Unlike current benchmarking or dynamic interaction paradigms, Agent-as-Interviewer utilizes agents to invoke knowledge tools for wider and deeper knowledge in the dynamic multi-turn question generation.<n>We develop JudgeAgent, a knowledge-wise dynamic evaluation framework that employs knowledge-driven synthesis as the agent's tool and uses difficulty scoring as strategy guidance.
  arXiv  Detail & Related papers  (2025-09-02T08:52:16Z)
• Teaching Language Models To Gather Information Proactively [53.85419549904644]
  Large language models (LLMs) are increasingly expected to function as collaborative partners.<n>In this work, we introduce a new task paradigm: proactive information gathering.<n>We design a scalable framework that generates partially specified, real-world tasks, masking key information.<n>Within this setup, our core innovation is a reinforcement finetuning strategy that rewards questions that elicit genuinely new, implicit user information.
  arXiv  Detail & Related papers  (2025-07-28T23:50:09Z)
• ReMA: Learning to Meta-think for LLMs with Multi-Agent Reinforcement Learning [53.817538122688944]
  We introduce Reinforced Meta-thinking Agents (ReMA) to elicit meta-thinking behaviors from Reasoning of Large Language Models (LLMs)<n>ReMA decouples the reasoning process into two hierarchical agents: a high-level meta-thinking agent responsible for generating strategic oversight and plans, and a low-level reasoning agent for detailed executions.<n> Empirical results from single-turn experiments demonstrate that ReMA outperforms single-agent RL baselines on complex reasoning tasks.
  arXiv  Detail & Related papers  (2025-03-12T16:05:31Z)
• Positive Experience Reflection for Agents in Interactive Text Environments [9.982616173090264]
  We introduce Sweet&Sour, a novel approach that incorporates positive experiences and managed memory to enrich the context available to the agent at decision time. Our comprehensive analysis spans both closed- and open-source LLMs and demonstrates the effectiveness of Sweet&Sour in improving agent performance.
  arXiv  Detail & Related papers  (2024-11-04T16:15:28Z)
• Self-Supervised Discovering of Interpretable Features for Reinforcement Learning [40.52278913726904]
  We propose a self-supervised interpretable framework for deep reinforcement learning. A self-supervised interpretable network (SSINet) is employed to produce fine-grained attention masks for highlighting task-relevant information. We verify and evaluate our method on several Atari 2600 games as well as Duckietown, which is a challenging self-driving car simulator environment.
  arXiv  Detail & Related papers  (2020-03-16T08:26:17Z)

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.