Research map / Probabilistic and scientific modeling / Molecules and proteins
Confusions over Time: An Interpretable Bayesian Model to Characterize Trends in Decision Making
Himabindu Lakkaraju, Jure Leskovec
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add temporal component (NT-CoT -> CoT) (ablation) | CoT · Bail, predicting item labels | AUC ROC: 0.65 -> 0.71 | Improved |
| add temporal component (NT-CoT -> CoT) (ablation) | CoT · Asthma, predicting item labels | AUC ROC: 0.68 -> 0.72 | Improved |
| add temporal component (NT-CoT -> CoT) (ablation) | CoT · Insurance, predicting item labels | AUC ROC: 0.69 -> 0.74 | Improved |
| add temporal component (NT-CoT -> CoT) (ablation) | CoT · Bail, inferring confusion matrices | MAE: 0.24 -> 0.19 | Improved |
| add temporal component (NT-CoT -> CoT) (ablation) | CoT · Asthma, predicting decisions | AUC ROC: 0.68 -> 0.74 | Improved |
| add interpretability components (prototypes, subspace indicators; NI-CoT -> CoT) (ablation) | CoT · Bail, predicting item labels | AUC ROC: 0.69 -> 0.71 | Improved |
| add interpretability components (NI-CoT -> CoT) (ablation) | CoT · Asthma, inferring confusion matrices | MAE: 0.18 -> 0.16 | Improved |
| add interpretability components (NI-CoT -> CoT) (ablation) | CoT · Insurance, predicting decisions | AUC ROC: 0.68 -> 0.71 | Improved |
| add interpretability components (NI-CoT -> CoT) (ablation) | CoT · Bail, cluster purity | purity: 0.72 -> 0.83 | Improved |
| add interpretability components (NI-CoT -> CoT) (ablation) | CoT · Bail, cluster inverse purity | inverse purity: 0.67 -> 0.78 | Improved |
| add temporal component (NT-CoT -> CoT) (ablation) | CoT · Insurance, cluster purity | purity: 0.79 -> 0.81 | Improved |
| replace JC baseline with CoT | JC · Bail, inferring confusion matrices | MAE: 0.26 -> 0.19 | Improved |
Closest papers by meaning
- Dissenting Explanations: Leveraging Disagreement to Reduce Model Overreliance (AAAI 2024 · similarity 0.84)
- Monitoring the Internal Monologue: Probe Trajectories Reveal Reasoning Dynamics (NeurIPS 2026 · similarity 0.82)
- Learning Deployable Causal Action Geometry under Temporal Non-Stationarity (NeurIPS 2026 · similarity 0.82)
- Thought Branches: Interpreting LLM Reasoning Requires Resampling (ICLR 2026 · similarity 0.82)
- Tracing the Traces: Latent Temporal Signals for Efficient and Accurate Reasoning (ICLR 2026 · similarity 0.82)
- Logic-Logit: A Logic-Based Approach to Choice Modeling (ICLR 2025 · similarity 0.82)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
