Papers › Designing a Dashboard for Transparency and Control of Conversational AI

Designing a Dashboard for Transparency and Control of Conversational AI

12 Jun 2024arXiv:2406.07882archive 2025-07-28

Yida Chen, Aoyu Wu, Trevor DePodesta, Catherine Yeh, Kenneth Li, Nicholas Castillo Marin, Oam Patel, Jan Riecke, Shivam Raval, Olivia Seow, Martin Wattenberg, Fernanda Viégas

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

PaperPDFCodeCode Syntology ran

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, on Syntology's MCP service (how to connect):

Code

Syntology 25 code samples harvested from 1 repository linked to this paper: 21 ran (of which 0 constructed an object rather than computing a result; 14 with no instrument failure: 0 honoured, 1 violated, 13 with no contract checked; 7 where Syntology's instrument failed) · 4 unverified. Where Syntology's instrument failed: 7 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: official repository: 25 samples from 1 repository, 21 ran (14 with no instrument failure; 7 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

Levels written on 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

yc015/talktuner-chatbot-llm-dashboard officialmentioned in papermentioned on GitHubpytorchMITSyntology: 25 samples, 21 ran (7 instrument-failed) · 4 unverified report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

25 samples harvested: 21 ran (of which 0 constructed an object rather than computing a result; 14 with no instrument failure: 0 honoured, 1 violated, 13 with no contract checked; 7 where Syntology's instrument failed) · 4 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

1ran · violated contract
7ran · our draft was wrong
13ran
4unverified

Licence: 0 of the 25 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from yc015/talktuner-chatbot-llm-dashboard. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of six reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity.

yc015/talktuner-chatbot-llm-dashboard official repository 25 samples, 21 ran (7 instrument-failed) · 4 unverified
async_post_service dashboard_v1/dashboard_task_queue/worker/worker.py ran MIT (permissive) · c65af88d2f239ac4 · report fixture: signature url:str/0/str, data:dict/0/? · returned coroutine · level recorded 2026-09-24
exp_evidence in 6 papers src/losses.py ran · our draft was wrong fingerprinted MIT (permissive) · ddba7773fd3d51f7 · report fixture: signature y:Tensor/2/float · returned Tensor · level recorded 2026-09-24
extract_activations_control dashboard_v1/probing/train_probes.py ran MIT (permissive) · dd58a8d833cc8b6c · report fixture: signature conversation:list/1/dict, model:callable/0/?, tokenizer:callable/0/?, model_type:str/0/str, num_layers:int/0/int, device:str/0/str · returned Tensor · level recorded 2026-09-24
format_conversation_for_model dashboard_v1/probing/train_probes.py ran MIT (permissive) · c9161e4fd0894771 · report fixture: signature conversation:list/1/dict, tokenizer:MockTokenizer/0/?, model_type:str/0/str · returned str · level recorded 2026-09-24
generate_conversation dashboard_v1/probing/generate_conversation_dataset.py ran MIT (permissive) · 5d58ce568acbe617 · report fixture: signature client:callable/0/?, attribute:str/0/str, target:str/0/str, is_non_attribute:bool/0/bool, system_prompt_template:NoneType/0/? · returned list · level recorded 2026-09-24
generate_dataset in 41 papers dashboard_v1/probing/generate_conversation_dataset.py ran MIT (permissive) · bc6d4f97d42f9501 · report fixture: signature api_key:str/0/str, attribute1:str/0/str, attribute2:str/0/str, target:str/0/str, num_conversations:int/0/int, meta_attribute:str/0/str, system_prompt_template:NoneType/0/? · returned dict · level recorded 2026-09-24
list_available_datasets dashboard_v1/probing/inspect_dataset.py ran · our draft was wrong fingerprinted MIT (permissive) · 79a8e24307cba632 · report fixture: signature base_dir:str/0/str · returned list · level recorded 2026-09-24
llama_v2_prompt src/dataset.py ran MIT (permissive) · a8ab0f29b2f5cd13 · report fixture: signature messages:list/1/dict, system_prompt:str/0/str · returned str · level recorded 2026-09-24
llama_v2_reverse src/prompt_utils.py ran · our draft was wrong MIT (permissive) · 01f8e1cc8000adb6 · report fixture: signature prompt:str/0/str · returned list · level recorded 2026-09-24
load_dataset in 276 papers dashboard_v1/probing/inspect_dataset.py ran · our draft was wrong MIT (permissive) · 01ff948df86446e4 · report fixture: signature filepath:str/0/str · returned dict · level recorded 2026-09-24
load_probe_classifier src/intervention_utils.py ran MIT (permissive) · afc1a18cc3093f96 · report fixture: signature model_func:callable/0/?, input_dim:int/0/int, num_classes:dict/0/?, weight_path:str/0/str · returned ProbeClassification · level recorded 2026-09-24
one_hot_embedding in 14 papers src/losses.py ran MIT (permissive) · d018404cd06ecc3f · report fixture: signature labels:Tensor/1/int, num_classes:int/0/int · returned Tensor · level recorded 2026-09-24
process_chat dashboard_v1/dashboard_task_queue/worker/worker.py ran MIT (permissive) · be819d515c7481ad · report fixture: signature task:dict/0/? · returned coroutine · level recorded 2026-09-24
process_chat_batched dashboard_v1/dashboard_task_queue/worker/worker.py ran MIT (permissive) · d6490eaa408ffc13 · report fixture: signature tasks:list/1/dict · returned coroutine · level recorded 2026-09-24
relu_evidence src/losses.py ran · our draft was wrong fingerprinted MIT (permissive) · efecc687b786a0cc · report fixture: signature y:Tensor/2/float · returned Tensor · level recorded 2026-09-24
remove_last_k_words src/dataset.py ran · our draft was wrong fingerprinted MIT (permissive) · c90fd1d879571fb6 · report fixture: signature s:str/0/str, k:int/0/int · returned str · level recorded 2026-09-24
return_classifier_dict src/intervention_utils.py ran MIT (permissive) · ed57d60ed6258a97 · report fixture: signature directory:str/0/str, model_func:callable/0/?, chosen_layer:NoneType/0/?, mix_scaler:bool/0/bool, sklearn:bool/0/bool · returned dict · level recorded 2026-09-24
split_conversation src/dataset.py ran MIT (permissive) · 764ffaea8c47e018 · report fixture: signature text:str/0/str, user_identifier:str/0/str, ai_identifier:str/0/str · returned tuple · level recorded 2026-09-24
split_into_messages src/intervention_utils.py ran · our draft was wrong fingerprinted MIT (permissive) · a6afc594b6f4973c · report fixture: signature text:str/0/str · returned list · level recorded 2026-09-24
start_probe_training dashboard_v1/probing/client_example.py ran MIT (permissive) · 5d1a14b83de26cdc · report fixture: signature api_url:str/0/str, openai_api_key:str/0/str, model:str/0/str, attribute1:str/0/str, target:str/0/str, num_conversations:int/0/int, attribute2:str/0/str, probe_type:str/0/str · returned tuple · level recorded 2026-09-24
validate_dataset dashboard_v1/probing/inspect_dataset.py ran · violated contract MIT (permissive) · 84df9b0e3fc9b985 · report fixture: signature dataset:dict/0/? · returned bool · level recorded 2026-09-24
get_task_status dashboard_v1/probing/client_example.py unverified MIT (permissive) · 93c779c76b35a006 · report
load_model_and_tokenizer in 61 papers dashboard_v1/probing/train_probes.py unverified MIT (permissive) · d0a8ad2ed431e1b5 · report
save_dataset dashboard_v1/probing/generate_conversation_dataset.py unverified MIT (permissive) · 73316eb27afceef3 · report
wait_for_task_completion dashboard_v1/probing/client_example.py unverified MIT (permissive) · e12290fa324335e9 · report

Tasks

Chatbot

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections