Loading the SOTA2 catalog…
SOTA2 Research · tasks
Explore the problems researchers are working on and the benchmarks used to measure progress.
| Task Name | Domain | Benchmarks | Papers | Last Update |
|---|---|---|---|---|
| Question Answering Correctness | 1 | 1 | Feb 18, 2026 | |
| Empathy assessment | 1 | 1 | Mar 9, 2026 | |
| Question Answering/Reasoning | 1 | 1 | May 18, 2026 | |
| End-to-end Time-to-First-Token (TTFT) | 1 | 1 | Mar 9, 2026 | |
| Visual Tic-Tac-Toe | 1 | 1 | May 18, 2026 | |
| Service Dialogue Task | 1 | 1 | Feb 18, 2026 | |
| Prediction-grounded correlation with output difference (JSD) | 1 | 1 | May 18, 2026 | |
| Correlation to Model Behavior Differences | 1 | 1 | May 18, 2026 | |
| Long-context text understanding |
| 1 |
| 1 |
| May 28, 2026 |
| Frontier Discovery and Ambiguity Disambiguation | 1 | 1 | May 18, 2026 |
|---|
| Long-context conversation reasoning | 1 | 1 | May 18, 2026 |
|---|
| Hallucination Steering | 1 | 1 | Mar 9, 2026 |
|---|
| Scientific research reasoning | 1 | 1 | May 18, 2026 |
|---|
| Open-ended generation (Medical Diagnostic QA) | 1 | 1 | May 28, 2026 |
|---|
| Counseling Response Generation | 1 | 1 | Feb 19, 2026 |
|---|
| Emotional Dialogue Generation | 1 | 1 | Mar 10, 2026 |
|---|
| Non-toxic generation | 1 | 1 | May 19, 2026 |
|---|
| Streaming Continual Visual-Instruction Tuning | 1 | 1 | May 19, 2026 |
|---|
| General Broad Information Seeking | 1 | 1 | Feb 18, 2026 |
|---|
| Solution-Focused Brief Therapy (SFBT) | 1 | 1 | Mar 10, 2026 |
|---|
| Safety Elicitation via Prompt and Query Refinement | 1 | 1 | May 19, 2026 |
|---|
| Conversational Quality Evaluation | 1 | 1 | May 19, 2026 |
|---|
| Math Proof Reward Modeling | 1 | 1 | Feb 18, 2026 |
|---|
| Turn-level correlation with human ratings | 1 | 1 | May 19, 2026 |
|---|
| Structured Commitment Generation | 1 | 1 | May 19, 2026 |
|---|