DRACULA: Hunting for the Actions Users Want Deep Research Agents to Execute

• URL: http://arxiv.org/abs/2604.23815v1
• Date: Sun, 26 Apr 2026 17:32:39 GMT
• Title: DRACULA: Hunting for the Actions Users Want Deep Research Agents to Execute
• Abstract summary: We collect DRACULA, the first dataset with user feedback on intermediate actions for Scientific Deep Research (DR) agents.<n>Our users select actions they prefer, then judge whether an output report applied their selections successfully.<n>We study the predictability of user-preferred actions via simulation-how well LLMs predict the actions users select-a step toward learning to generate useful actions.
• Score: 66.25314396197075
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Scientific Deep Research (DR) agents answer user queries by synthesizing research papers into multi-section reports. User feedback can improve their utility, but existing protocols only score the final report, making it hard to study and learn which intermediate actions DR agents should take to improve reports. We collect DRACULA, the first dataset with user feedback on intermediate actions for DR. Over five weeks, nineteen expert CS researchers ask queries to a DR system that proposes actions (e.g., "Add a section on datasets"). Our users select actions they prefer, then judge whether an output report applied their selections successfully, yielding 8,103 action preferences and 5,230 execution judgments. After confirming a DR agent can execute DRACULA's actions, we study the predictability of user-preferred actions via simulation-how well LLMs predict the actions users select-a step toward learning to generate useful actions. We discover: (1) LLM judges initially struggle to predict action selections, but improve most when using a user's full selection history, rather than self-reported or extrapolated user context signals; (2) Users' selections for the same query differ based on unstated goals, bottlenecking simulation and motivating affordances that let users steer reports; and (3) Our simulation results inform an online intervention that generates new actions based on the user's past interactions, which users pick most often in follow-up studies. Overall, while work extensively studies execution, DRACULA reveals a key challenge is deciding which actions to execute in the first place. We open-source DRACULA's study design, user feedback, and simulation tasks to spur future work on action feedback for long-horizon agents.

Related papers

• SWE-Together: Evaluating Coding Agents in Interactive User Sessions [19.069719245971786]
  We introduce SWE-Together, a multi-turn benchmark reconstructed from real user-agent coding sessions.<n>We curate 109 repository-level tasks from 11,260 recorded sessions, selecting sessions with recoverable repository states, clear user goals, and observable outcomes.<n>We build a reactive LLM-based user simulator that preserves the original users' intents and provides feedback when the coding agent's progress requires it.
  arXiv  Detail & Related papers  (2026-06-29T08:35:15Z)
• IntentRL: Training Proactive User-intent Agents for Open-ended Deep Research via Reinforcement Learning [54.21689544323704]
  Deep Research (DR) agents extend Large Language Models (LLMs) beyond parametric knowledge.<n>Unlike real-time conversational assistants, DR is computationally expensive and time-consuming.<n>We propose IntentRL, a framework that trains proactive agents to clarify latent user intents before starting long-horizon research.
  arXiv  Detail & Related papers  (2026-02-03T12:43:09Z)
• Search-Based Interaction For Conversation Recommendation via Generative Reward Model Based Simulated User [117.82681846559909]
  Conversational recommendation systems (CRSs) use multi-turn interaction to capture user preferences and provide personalized recommendations.<n>We propose a generative reward model based simulated user, named GRSU, for automatic interaction with CRSs.
  arXiv  Detail & Related papers  (2025-04-29T06:37:30Z)
• Bidirectional Decoding: Improving Action Chunking via Guided Test-Time Sampling [51.38330727868982]
  We show how action chunking impacts the divergence between a learner and a demonstrator.<n>We propose Bidirectional Decoding (BID), a test-time inference algorithm that bridges action chunking with closed-loop adaptation.<n>Our method boosts the performance of two state-of-the-art generative policies across seven simulation benchmarks and two real-world tasks.
  arXiv  Detail & Related papers  (2024-08-30T15:39:34Z)
• How Reliable is Your Simulator? Analysis on the Limitations of Current LLM-based User Simulators for Conversational Recommendation [14.646529557978512]
  We analyze the limitations of using Large Language Models in constructing user simulators for Conversational Recommender System. Data leakage, which occurs in conversational history and the user simulator's replies, results in inflated evaluation results. We propose SimpleUserSim, employing a straightforward strategy to guide the topic toward the target items.
  arXiv  Detail & Related papers  (2024-03-25T04:21:06Z)
• On Generative Agents in Recommendation [58.42840923200071]
  Agent4Rec is a user simulator in recommendation based on Large Language Models. Each agent interacts with personalized recommender models in a page-by-page manner.
  arXiv  Detail & Related papers  (2023-10-16T06:41:16Z)
• AgentCF: Collaborative Learning with Autonomous Language Agents for Recommender Systems [112.76941157194544]
  We propose AgentCF for simulating user-item interactions in recommender systems through agent-based collaborative filtering. We creatively consider not only users but also items as agents, and develop a collaborative learning approach that optimize both kinds of agents together. Overall, the optimized agents exhibit diverse interaction behaviors within our framework, including user-item, user-user, item-item, and collective interactions.
  arXiv  Detail & Related papers  (2023-10-13T16:37:14Z)
• Explainable Active Learning for Preference Elicitation [0.0]
  We employ Active Learning (AL) to solve the addressed problem with the objective of maximizing information acquisition with minimal user effort. AL operates for selecting informative data from a large unlabeled set to inquire an oracle to label them. It harvests user feedback (given for the system's explanations on the presented items) over informative samples to update an underlying machine learning (ML) model.
  arXiv  Detail & Related papers  (2023-09-01T09:22:33Z)
• Personalized Algorithmic Recourse with Preference Elicitation [20.78332455864586]
  We introduce PEAR, the first human-in-the-loop approach capable of providing personalized algorithmic recourse tailored to the needs of any end-user. PEAR builds on insights from Bayesian Preference Elicitation to iteratively refine an estimate of the costs of actions by asking choice set queries to the target user. Our empirical evaluation on real-world datasets highlights how PEAR produces high-quality personalized recourse in only a handful of iterations.
  arXiv  Detail & Related papers  (2022-05-27T03:12:18Z)
• RETE: Retrieval-Enhanced Temporal Event Forecasting on Unified Query Product Evolutionary Graph [18.826901341496143]
  Temporal event forecasting is a new user behavior prediction task in a unified query product evolutionary graph. We propose a novel RetrievalEnhanced Event forecasting framework. Unlike existing methods, we propose methods that enhance user representations via roughly connected entities in the whole graph.
  arXiv  Detail & Related papers  (2022-02-12T19:27:56Z)

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.