Paper Arena
Compare research contributions, evidence and reusable work on a six-tier rubric. Scores come from an independent review.
Review snapshot Scoring rubric v1.0.0 ·
Papers by tier
Reviewed 21Paper leaderboard
Reviewed 21·Awaiting a rating 1Ranks and ties follow the review record. TBD marks a title awaiting a match to the site and may include work under review or awaiting submission. Papers marked “under view” are in scope and await an arena rating.
| Rank | Paper | Score | Tier | Work date |
|---|---|---|---|---|
| 1 | TBD | 8.5 | Tier 2 | Date pending |
| 2 | TBD | 8.5 | Tier 2 | Date pending |
| 3 | TBD | 8.5 | Tier 2 | Date pending |
| 4 | TBD | 8.5 | Tier 2 | Date pending |
| 5 | TBD | 8.0 | Tier 2 | Date pending |
| 6 | When Is Program Structure Required? Certificate-Grounded Calibration of Program-Induction Difficulty arXiv 2026 | 8.0 | Tier 2 | Preprint posted |
| 7 | Complex Utility, Simple Calibration: Decision Sufficiency in RAG Evidence Selection WSDM 2027 | 8.0 | Tier 2 | Accepted |
| 8 | TBD | 8.0 | Tier 2 | Date pending |
| 9 | TBD | 8.0 | Tier 2 | Date pending |
| 10 | TBD | 8.0 | Tier 2 | Date pending |
| 11 | TBD | 7.5 | Tier 3 | Date pending |
| 12 | TBD | 7.5 | Tier 3 | Date pending |
| 13 | TBD | 7.5 | Tier 3 | Date pending |
| 14 | TBD | 7.5 | Tier 3 | Date pending |
| 15 | TBD | 7.5 | Tier 3 | Date pending |
| 16 | TBD | 7.0 | Tier 3 | Date pending |
| 17 | TBD | 7.0 | Tier 3 | Date pending |
| 18 | PAIRSAFE: Proxy-Resistant Paired Safety Evaluation for Refusal Boundaries AACL-IJCNLP 2026 | 6.5 | Tier 4 | Accepted |
| 19 | Contract-Governed Training for Earth Observation: Observed Service Agreement Graphs and Coverage-Accuracy Trade-offs IGARSS 2026 | 6.0 | Tier 4 | Accepted |
| 20 | Optimistic Feasible Search for Closed-Loop Fair Threshold Decision-Making WCCI (IJCNN) 2026 | 5.5 | Tier 4 | Accepted |
| 21 | When +1% Is Not Enough: A Paired Bootstrap Protocol for Evaluating Small Improvements PAKDD 2026 | Qualitative | Tier 5 | Accepted |
| — | ARC-CTRL: Audit-Guided Closed-Loop Trust Control for Surrogate-Assisted Evolutionary Hyperparameter Optimization WCCI (CEC) 2026 | under view | — | Accepted |
No entries match these filters.
Date order uses the date shown for each paper. Entries with a pending date appear last.
Scoring rubricv1.0.0 ·
| Tier | Score | Interpretation |
|---|---|---|
| Tier 1 | 9–10 | The work offers an important new insight or gives researchers a capability they previously lacked. Strong evidence covers the key conditions under which the central claim holds. The contribution is substantial enough to change how researchers understand or approach the problem. |
| Tier 2 | 8–8.5 | The work makes a clear and important advance over prior research. Ample evidence supports the central conclusion, and the paper gives convincing answers to the main questions that could change it. |
| Tier 3 | 7–7.5 | The work solves a well-defined research problem and makes a complete contribution worth sharing. The available evidence supports the conclusions under the stated conditions, and the limits of their applicability are clear. |
| Tier 4 | 5–6.5 | The work makes an identifiable contribution, but gaps remain in key comparisons, validation or reasoning. These gaps affect readers’ ability to judge whether the main conclusions hold or under what conditions they generalize. |
| Tier 5 | Qualitative | The main value lies in reusable data, code, experimental records or practical experience. The material is suitable for a technical report or an open resource that other researchers can consult and use. |
| Tier 6 | Qualitative | A clear independent contribution has yet to emerge, and the material offers limited results that others can reuse. It is suitable for archiving as a record of the research process. |