Skip to content
← Back to explorer

HFEPX Benchmark Hub

LMSYS Chatbot Arena Or AIME Benchmark Papers

Updated from current HFEPX corpus (Mar 21, 2026). 21 papers are grouped in this benchmark page.

Read Full Context

Updated from current HFEPX corpus (Mar 21, 2026). 21 papers are grouped in this benchmark page. Common evaluation modes: Automatic Metrics, Human Eval. Common annotation unit: Pairwise. Frequent quality control: Calibration. Frequently cited benchmark: LMSYS Chatbot Arena. Common metric signal: pass@1. Use this page to compare protocol setup, judge behavior, and labeling design decisions before running new eval experiments. Newest paper in this set is from Feb 13, 2026.

Papers: 21 Last published: Feb 13, 2026 Global RSS

Researcher Quick Triage

Use this page for benchmark-matched method comparisons and eval protocol selection. Quality band: Medium .

High-Signal Coverage

100.0%

21 / 21 sampled papers are not low-signal flagged.

Replication-Ready Set

4

Papers with explicit benchmark + metric + eval mode fields.

Quality Controls

4.8%

1 papers report calibration/adjudication/IAA controls.

  • 11 papers explicitly name benchmark datasets in the sampled set.
  • 4 papers report at least one metric term in metadata extraction.
  • Start with the ranked shortlist below before reading all papers.

Primary action: Start with the top 2 benchmark-matched papers, then compare evaluation modes in the protocol matrix.

Why This Matters (Expanded)

Why This Matters For Eval Research

  • 100% of papers report explicit human-feedback signals, led by pairwise preferences.
  • automatic metrics appears in 19% of papers in this hub.
  • LMSYS Chatbot Arena is a recurring benchmark anchor for cross-paper comparisons in this page.
Protocol Notes (Expanded)

Protocol Takeaways

  • Most common quality-control signal is rater calibration (4.8% of papers).
  • Rater context is mostly unspecified rater pools, and annotation is commonly pairwise annotation; use this to scope replication staffing.
  • Compare papers that report both human_eval and llm_as_judge to quantify judge-human agreement drift.

Benchmark Interpretation

  • LMSYS Chatbot Arena appears in 63.6% of hub papers (7/21); use this cohort for benchmark-matched comparisons.
  • AIME appears in 36.4% of hub papers (4/21); use this cohort for benchmark-matched comparisons.

Metric Interpretation

  • pass@1 is reported in 18.2% of hub papers (2/21); compare with a secondary metric before ranking methods.
  • accuracy is reported in 9.1% of hub papers (1/21); compare with a secondary metric before ranking methods.

Start Here (Benchmark-Matched First 6)

Ranked by protocol completeness so you can quickly find papers suitable for comparison studies.

Protocol Matrix (Top 10)

Compare protocol ingredients quickly before deep-reading full papers.

Paper Eval Modes Human Feedback Metrics Quality Controls
SCOPE: Selective Conformal Optimized Pairwise LLM Judging

Feb 13, 2026

Automatic Metrics Pairwise Preference Error rate Calibration
$V_1$: Unifying Generation and Self-Verification for Parallel Reasoners

Mar 4, 2026

Automatic Metrics Pairwise Preference Pass@1 Not reported
Don't Pass@k: A Bayesian Framework for Large Language Model Evaluation

Oct 5, 2025

Automatic Metrics, Simulation Env Rubric Rating Accuracy, Pass@k Not reported
Critique-GRPO: Advancing LLM Reasoning with Natural Language and Numerical Feedback

Jun 3, 2025

Automatic Metrics Critique Edit Pass@1 Not reported
Think$^{2}$: Grounded Metacognitive Reasoning in Large Language Models

Feb 21, 2026

Human Eval Pairwise Preference Not reported Not reported
Examining Reasoning LLMs-as-Judges in Non-Verifiable LLM Post-Training

Mar 12, 2026

Not reported Pairwise Preference Not reported Not reported
WebCoderBench: Benchmarking Web Application Generation with Comprehensive and Interpretable Evaluation Metrics

Jan 5, 2026

Llm As Judge Pairwise Preference Not reported Not reported
Revisiting Self-Play Preference Optimization: On the Role of Prompt Difficulty

Oct 7, 2025

Not reported Pairwise Preference Not reported Not reported
Alignment through Meta-Weighted Online Sampling: Bridging the Gap between Data Generation and Preference Optimization

Sep 27, 2025

Not reported Pairwise Preference Not reported Not reported
A Third Paradigm for LLM Evaluation: Dialogue Game-Based Evaluation using clembench

Jul 11, 2025

Not reported Pairwise Preference Not reported Not reported
Researcher Workflow (Detailed)

Checklist

  • Strong: Papers with explicit human feedback

    Coverage is strong (100% vs 45% target).

  • Gap: Papers reporting quality controls

    Coverage is a replication risk (9.1% vs 30% target).

  • Strong: Papers naming benchmarks/datasets

    Coverage is strong (100% vs 35% target).

  • Strong: Papers naming evaluation metrics

    Coverage is strong (36.4% vs 35% target).

  • Gap: Papers with known rater population

    Coverage is a replication risk (0% vs 35% target).

  • Strong: Papers with known annotation unit

    Coverage is strong (36.4% vs 35% target).

Strengths

  • Strong human-feedback signal (100% of papers).
  • Most papers provide measurable evaluation context (100% benchmarks, 36.4% metrics).
  • Contains both human-eval and LLM-as-judge protocols for head-to-head methodology comparison.

Known Gaps

  • Only 9.1% of papers report quality controls; prioritize calibration/adjudication evidence.
  • Rater population is under-specified (0% coverage).
  • LLM-as-judge appears without enough inter-annotator agreement reporting.

Suggested Next Analyses

  • Compare papers that report both human_eval and llm_as_judge to quantify judge-human agreement drift.
  • Stratify by benchmark (LMSYS Chatbot Arena vs AIME) before comparing methods.
  • Track metric sensitivity by reporting both pass@1 and accuracy.
  • Add inter-annotator agreement checks when reproducing these protocols.

Recommended Queries

Known Limitations
  • Only 9.1% of papers report quality controls; prioritize calibration/adjudication evidence.
  • Rater population is under-specified (0% coverage).
  • Narrative synthesis is grounded in metadata and abstracts only; full-paper implementation details are not parsed.
Research Utility Snapshot (Detailed)

Evaluation Modes

  • Automatic Metrics (4)
  • Human Eval (1)
  • Llm As Judge (1)
  • Simulation Env (1)

Human Feedback Mix

  • Pairwise Preference (9)
  • Critique Edit (1)
  • Rubric Rating (1)

Top Benchmarks

  • LMSYS Chatbot Arena (7)
  • AIME (4)
  • Arena Hard (3)
  • AlpacaEval (2)

Top Metrics

  • Pass@1 (2)
  • Accuracy (1)
  • Error rate (1)
  • Pass@k (1)

Top Papers On This Benchmark

Related Benchmark Hubs

Need human evaluators for your AI research? Scale annotation with expert AI Trainers.