mlsanityData to 7 Oct 2026

Designing a Dashboard for Transparency and Control of Conversational AI

arXiv 2406.07882 ↗First posted 12 Jun 2024cs.CL cs.AI cs.HCcited by 43

Not on the map. Why?

Abstract (arXiv)

Conversational LLMs function as black box systems, leaving users guessing about why they see the output they do. This lack of transparency is potentially problematic, especially given concerns around bias and truthfulness. To address this issue, we present an end-to-end prototype-connecting interpretability techniques with user experience design-that seeks to make chatbots more transparent. We begin by showing evidence that a prominent open-source LLM has a "user model": examining the internal state of the system, we can extract data related to a user's age, gender, educational level, and socioeconomic status. Next, we describe the design of a dashboard that accompanies the chatbot interface, displaying this user model in real time. The dashboard can also be used to control the user model and the system's behavior. Finally, we discuss a study in which users conversed with the instrumented system. Our results suggest that users appreciate seeing internal states, which helped them expose biased behavior and increased their sense of control. Participants also made valuable suggestions that point to future directions for both design and machine learning research. The project page and video demo of our TalkTuner system are available at https://bit.ly/talktuner-project-page

Key facts. Cited by 43 papers in the map (data through 7 Oct 2026). Builds on 26 of its 54 references that are in the map. First posted on arXiv 12 Jun 2024.

Builds on

26 of 54 references.

References · 26

  1. 2023GPT-4 Technical Reportcited by 18,132
  2. 2023Llama 2: Open Foundation and Fine-Tuned Chat Modelscited by 12,865
  3. 2023Judging LLM-as-a-Judge with MT-Bench and Chatbot Arenacited by 8,393
  4. 2017Interpretability Beyond Feature Attribution: Quantitative Testing with Concept Activation Vectors (TCAV)cited by 1,309
  5. 2016Understanding intermediate layers using linear classifier probescited by 1,279
  6. 2023Inference-Time Intervention: Eliciting Truthful Answers from a Language Modelcited by 1,259
  7. 2023Representation Engineering: A Top-Down Approach to AI Transparencycited by 1,239
  8. 2024Chatbot Arena: An Open Platform for Evaluating LLMs by Human Preferencecited by 1,151
  9. 2023Language Models Don't Always Say What They Think: Unfaithful Explanations in Chain-of-Thought Promptingcited by 1,081
  10. 2023ChatGPT Outperforms Crowd-Workers for Text-Annotation Taskscited by 849
  11. 2021Probing Classifiers: Promises, Shortcomings, and Advancescited by 831
  12. 2023Steering Language Models With Activation Engineeringcited by 824
14 more
  1. 2022Discovering Language Model Behaviors with Model-Written Evaluationscited by 805
  2. 2023The Geometry of Truth: Emergent Linear Structure in Large Language Model Representations of True/False Datasetscited by 673
  3. 2022SODA: Million-scale Dialogue Distillation with Social Commonsense Contextualizationcited by 166
  4. 2023MathDial: A Dialogue Tutoring Dataset with Rich Pedagogical Properties Grounded in Math Reasoning Problemscited by 128
  5. 2019Human-Centered Tools for Coping with Imperfect Algorithms during Medical Decision-Makingcited by 109
  6. 2023Topical-Chat: Towards Knowledge-Grounded Open-Domain Conversationscited by 88
  7. 2023ChainForge: A Visual Toolkit for Prompt Engineering and LLM Hypothesis Testingcited by 88
  8. 2023Do Models Explain Themselves? Counterfactual Simulatability of Natural Language Explanationscited by 85
  9. 2023PLACES: Prompting Language Models for Social Conversation Synthesiscited by 72
  10. 2020PANDORA Talks: Personality and Demographics on Redditcited by 40
  11. 2024LLM Comparator: Visual Analytics for Side-by-Side Evaluation of Large Language Modelscited by 29
  12. 2023Bias and Fairness in Chatbots: An Overviewcited by 20
  13. 2023The System Model and the User Model: Exploring AI Dashboard Designcited by 12
  14. 2018Gender bias in academic recruitmentcited by 1

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 10; 2025: 19; 2026: 14.

43 papers.

Sort:

By area:

  1. 2024LatentQA: Teaching LLMs to Decode Activations Into Natural Languagecited by 49
  2. 2024WaitGPT: Monitoring and Steering Conversational LLM Agent in Data Analysis with On-the-Fly Code Visualizationcited by 38
  3. 2024Stereotype or Personalization? User Identity Biases Chatbot Recommendationscited by 33
  4. 2024Towards Unifying Interpretability and Control: Evaluation via Interventioncited by 28
  5. 2025Training Language Models to Explain Their Own Computationscited by 27
  6. 2024The Quest for the Right Mediator: Surveying Mechanistic Interpretability Through the Lens of Causal Mediation Analysiscited by 24
  7. 2025Reading Between the Prompts: How Stereotypes Shape LLM's Implicit Personalizationcited by 24
  8. 2025Belief Dynamics Reveal the Dual Nature of In-Context Learning and Activation Steeringcited by 24
  9. 2024Relational Composition in Neural Networks: A Survey and Call to Actioncited by 23
  10. 2024Proactive Agents for Multi-Turn Text-to-Image Generation Under Uncertaintycited by 23
  11. 2026Sycophancy Hides Linearly in the Attention Headscited by 21
  12. 2026Language Model Circuits Are Sparse in the Neuron Basiscited by 18
31 more
  1. 2025Robustly Improving LLM Fairness in Realistic Settings via Interpretabilitycited by 16
  2. 2024LLM Targeted Underperformance Disproportionately Impacts Vulnerable Userscited by 15
  3. 2026ODESteer: A Unified ODE-Based Steering Framework for LLM Alignmentcited by 15
  4. 2025Because we have LLMs, we Can and Should Pursue Agentic Interpretabilitycited by 14
  5. 2025The Geometry of Self-Verification in a Task-Specific Reasoning Modelcited by 12
  6. 2025Interpretation Meets Safety: A Survey on Interpretation Methods and Tools for Improving LLM Safetycited by 10
  7. 2025MONICA: Real-Time Monitoring and Calibration of Chain-of-Thought Sycophancy in Large Reasoning Modelscited by 10
  8. 2026Steering Vector Fields for Context-Aware Inference-Time Control in Large Language Modelscited by 10
  9. 2024ChatGPT Doesn't Trust Chargers Fans: Guardrail Sensitivity in Contextcited by 9
  10. 2026One Persona, Many Cues, Different Results: How Sociodemographic Cues Impact LLM Personalizationcited by 9
  11. 2025Language Models Change Facts Based on the Way You Talkcited by 8
  12. 2025Emergence of Hierarchical Emotion Organization in Large Language Modelscited by 6
  13. 2025Neologism Learning for Controllability and Self-Verbalizationcited by 5
  14. 2026Consistency Training while Mitigating Obfuscation via Rate Matchingcited by 4
  15. 2025Enhancing Conversational Agents with Theory of Mind: Aligning Beliefs, Desires, and Intentions for Human-Like Interactioncited by 3
  16. 2025Web-Browsing LLMs Can Access Social Media Profiles and Infer User Demographicscited by 3
  17. 2026Verbalizing LLMs' assumptions to explain and control sycophancycited by 3
  18. 2026Tracing Persona Vectors Through LLM Pretrainingcited by 3
  19. 2024Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systemscited by 2
  20. 2025Neural Transparency: Mechanistic Interpretability Interfaces for Anticipating Model Behaviors for Personalized AIcited by 2
  21. 2026Auditing Support Strategies in LLMs through Grounded Multi-Turn Social Simulationcited by 2
  22. 2025Regulating the Agency of LLM-based Agentscited by 1
  23. 2025Prioritize Economy or Climate Action? Investigating ChatGPT Response Differences Based on Inferred Political Orientationcited by 1
  24. 2025Decomposing Theory of Mind: How Emotional Processing Mediates ToM Abilities in LLMscited by 1
  25. 2026Topics as Proxies for Sociodemographics: How Conversational Context Affects LLM Answerscited by 1
  26. 2025Clones in the Machine: A Feminist Critique of Agency in Digital Cloningcited by 0
  27. 2025Fine-Grained Interpretation of Political Opinions in Large Language Modelscited by 0
  28. 2026The Fake Friend Dilemma: Relational Trust and the Political Economy of Conversational AIcited by 0
  29. 2026Mechanistic Indicators of Steering Effectiveness in Large Language Modelscited by 0
  30. 2026Tensor Product Representation Probes Reveal Shared Structure Across Linear Directionscited by 0
  31. 2026User Model Extraction via Belief Self-Distillationcited by 0