Detecting when Users Disagree with Generated Captions
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
The pervasive integration of artificial intelligence (AI) into daily life has led to a growing interest in AI agents that can learn continuously. Interactive Machine Learning (IML) has emerged as a promising approach to meet this need, essentially involving human experts in the model training process, often through iterative user feedback. However, repeated feedback requests can lead to frustration and reduced trust in the system. Hence, there is increasing interest in refining how these systems interact with users to ensure efficiency without compromising user experience. Our research investigates the potential of eye tracking data as an implicit feedback mechanism to detect user disagreement with AI-generated captions in image captioning systems. We conducted a study with 30 participants using a simulated captioning interface and gathered their eye movement data as they assessed caption accuracy. The goal of the study was to determine whether eye tracking data can predict user agreement or disagreement effectively, thereby strengthening IML frameworks. Our findings reveal that, while eye tracking shows promise as a valuable feedback source, ensuring consistent and reliable model performance across diverse users remains a challenge.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it