Skip to content
← Back to explorer

HFEPX Hub

CS.CL + Web Browsing Papers

Updated from current HFEPX corpus (Apr 12, 2026). 50 papers are grouped in this hub page.

Read Full Context

Updated from current HFEPX corpus (Apr 12, 2026). 50 papers are grouped in this hub page. Common evaluation modes: Automatic Metrics, Simulation Env. Most common rater population: Domain Experts. Common annotation unit: Trajectory. Frequent quality control: Adjudication. Frequently cited benchmark: WebArena. Common metric signal: accuracy. 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 Mar 22, 2026.

Papers: 50 Last published: Mar 22, 2026 Global RSS Tag RSS
Cs.CLWeb Browsing

Researcher Quick Triage

This hub is best used for protocol triage and replication planning from abstract-level evidence. Quality band: Medium .

All Sampled Papers (50) Replication-Ready Only (10)

High-Signal Coverage

100.0%

50 / 50 sampled papers are not low-signal flagged.

Replication-Ready Set

10

Benchmark + metric + eval mode explicitly present.

Judge/Human Comparability

1

Papers containing both `human_eval` and `llm_as_judge`.

  • 10 papers are replication-ready (benchmark + metric + explicit evaluation mode).
  • 1 papers support judge-vs-human agreement analysis.
  • 3 papers report explicit quality controls (calibration/adjudication/IAA).

Primary action: Start with the top 2 papers in “Start Here”, then validate assumptions in the protocol matrix.

Currently showing only replication-ready papers in ranking and matrix sections (10 papers).

Need evaluators for this research workflow?

Post a Job →

Why This Matters For Eval Research

  • 47.2% of papers report explicit human-feedback signals, led by pairwise preferences.
  • automatic metrics appears in 50% of papers in this hub.
  • WebArena is a recurring benchmark anchor for cross-paper comparisons in this page.

Protocol Takeaways

  • 1 sampled papers report both human evaluation and LLM-as-judge, supporting direct agreement checks.
  • Most common quality-control signal is adjudication (6% of papers).
  • Rater context is mostly domain experts, and annotation is commonly trajectory-level annotation; use this to scope replication staffing.

Benchmark Interpretation

  • WebArena appears in 11.1% of hub papers (4/50); use this cohort for benchmark-matched comparisons.
  • OSWorld appears in 5.6% of hub papers (2/50); use this cohort for benchmark-matched comparisons.

Metric Interpretation

  • accuracy is reported in 44.4% of hub papers (16/50); compare with a secondary metric before ranking methods.
  • cost is reported in 13.9% of hub papers (5/50); compare with a secondary metric before ranking methods.
Researcher Checklist (Expanded)

Researcher Checklist

  • Strong: Papers with explicit human feedback

    Coverage is strong (47.2% vs 45% target).

  • Gap: Papers reporting quality controls

    Coverage is a replication risk (8.3% vs 30% target).

  • Strong: Papers naming benchmarks/datasets

    Coverage is strong (44.4% vs 35% target).

  • Strong: Papers naming evaluation metrics

    Coverage is strong (75% vs 35% target).

  • Gap: Papers with known rater population

    Coverage is a replication risk (11.1% vs 35% target).

  • Moderate: Papers with known annotation unit

    Coverage is usable but incomplete (27.8% vs 35% target).

Strengths

  • Strong human-feedback signal (47.2% of papers).
  • Most papers provide measurable evaluation context (44.4% benchmarks, 75% metrics).
  • Contains both human-eval and LLM-as-judge protocols for head-to-head methodology comparison.

Known Gaps

  • Only 8.3% of papers report quality controls; prioritize calibration/adjudication evidence.
  • Rater population is under-specified (11.1% 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 (WebArena vs OSWorld) before comparing methods.
  • Track metric sensitivity by reporting both accuracy and cost.
  • Add inter-annotator agreement checks when reproducing these protocols.
Recommended Queries (Expanded)

Recommended Queries

Judge vs Human Agreement Benchmark Slice: WebArena Metric Slice: accuracy Recent High-Signal Papers

Start with These 3

Use these when you need one protocol anchor, one benchmark anchor, and one recent comparison point before reading the wider hub.

Start Here (Best First 6)

Ranked for protocol completeness (human signal, benchmark + metric anchors, quality controls, and judge/human overlap).

Protocol Matrix (Top 12)

Use this to quickly compare protocol ingredients instead of scanning long prose.

Paper HF Signal Eval Modes Benchmarks Metrics QC
AgentHER: Hindsight Experience Replay for LLM Agent Trajectory Relabeling

Mar 22, 2026

 Yes
Demonstrations
Human Eval , Llm As Judge WebArena , ToolBench Precision , Pass@1 Not Reported
SODIUM: From Open Web Data to Queryable Databases

Mar 19, 2026

 Yes
Expert Verification
Automatic Metrics Sodium Bench Accuracy Not Reported
MemoryArena: Benchmarking Agent Memory in Interdependent Multi-Session Agentic Tasks

Feb 18, 2026

 Yes
Pairwise Preference
Automatic Metrics Memoryarena Recall Not Reported
Let's Think in Two Steps: Mitigating Agreement Bias in MLLMs with Self-Grounded Verification

Jul 15, 2025

 Yes
Pairwise Preference
Automatic Metrics , Simulation Env VisualWebArena , OSWorld Accuracy Not Reported
RedTeamCUA: Realistic Adversarial Testing of Computer-Use Agents in Hybrid Web-OS Environments

May 28, 2025

 Yes
Red Team
Automatic Metrics Rtc Bench Jailbreak success rate Not Reported
LUDOBENCH: Evaluating LLM Behavioural Decision-Making Through Spot-Based Board Game Scenarios in Ludo

Apr 7, 2026

 No
Not Reported
Simulation Env Ludobench Dice Not Reported
Don't Overthink It: Inter-Rollout Action Agreement as a Free Adaptive-Compute Signal for LLM Agents

Apr 9, 2026

 No
Not Reported
Automatic Metrics GSM8K Accuracy Not Reported
SpatiaLab: Can Vision-Language Models Perform Spatial Reasoning in the Wild?

Feb 3, 2026

 No
Not Reported
Automatic Metrics DROP Accuracy Not Reported
Go-Browse: Training Web Agents with Structured Exploration

Jun 4, 2025

 No
Not Reported
Simulation Env WebArena Success rate Not Reported
Healthy LLMs? Benchmarking LLM Knowledge of UK Government Public Health Information

May 9, 2025

 No
Not Reported
Automatic Metrics Pubhealthbench Accuracy Not Reported

Protocol Diff (Top Papers)

Fast side-by-side comparison for the highest-ranked papers in this hub.

Signal AgentHER: Hindsight Experience Replay for LLM Agent… SODIUM: From Open Web Data to Queryable Databases MemoryArena: Benchmarking Agent Memory in Interdepe…
Human Feedback DemonstrationsExpert VerificationPairwise Preference
Evaluation Modes Human Eval, Llm As JudgeAutomatic MetricsAutomatic Metrics
Benchmarks WebArena, ToolBenchSodium BenchMemoryarena
Metrics Precision, Pass@1AccuracyRecall
Quality Controls Not reportedNot reportedNot reported
Rater Population UnknownDomain ExpertsUnknown
Annotation Unit TrajectoryUnknownUnknown
Suggested Reading Order (Extended)

This section is intentionally expanded only when needed; use “Start Here” above for a faster pass.

Suggested Reading Order

  1. CounselReflect: A Toolkit for Auditing Mental-Health Dialogues

    Start here for detailed protocol reporting and quality-control evidence. Signals: human evaluation + rubric ratings. Abstract: The system integrates two families of evaluation signals: (i) 12 model-based metrics.

  2. Don't Overthink It: Inter-Rollout Action Agreement as a Free Adaptive-Compute Signal for LLM Agents

    High citation traction makes this a strong baseline for protocol comparison. Signals: automatic metrics. Focus: GSM8K / accuracy. Abstract: Inference-time compute scaling has emerged as a powerful technique.

  3. AgentGL: Towards Agentic Graph Learning with LLMs via Reinforcement Learning

    High citation traction makes this a strong baseline for protocol comparison. Signals: automatic metrics. Focus: accuracy. Abstract: Large Language Models (LLMs) increasingly rely on agentic capabilities-iterative retrieval, tool.

  4. LUDOBENCH: Evaluating LLM Behavioural Decision-Making Through Spot-Based Board Game Scenarios in Ludo

    High citation traction makes this a strong baseline for protocol comparison. Signals: simulation environments. Focus: Ludobench / dice. Abstract: We introduce LudoBench, a benchmark for evaluating LLM strategic.

  5. AgentHER: Hindsight Experience Replay for LLM Agent Trajectory Relabeling

    Include a human-eval paper to calibrate against judge-based evaluation settings. Signals: human evaluation + demonstration data. Focus: WebArena / precision. Abstract: AgentHER realises this idea through a four-stage.

  6. InnoEval: On Research Idea Evaluation as a Knowledge-Grounded, Multi-Perspective Reasoning Problem

    Include an LLM-as-judge paper to test judge design and agreement assumptions. Signals: LLM-as-judge. Focus: Innoeval. Abstract: However, existing idea evaluation methods often suffer from narrow knowledge horizons, flattened.

  7. Let's Think in Two Steps: Mitigating Agreement Bias in MLLMs with Self-Grounded Verification

    Adds automatic metrics with pairwise preferences for broader protocol coverage within this hub. Signals: automatic metrics + pairwise preferences. Focus: VisualWebArena / accuracy. Abstract: Multimodal LLMs (MLLMs) offer.

  8. Vision-DeepResearch Benchmark: Rethinking Visual and Textual Search for Multimodal Large Language Models

    Adds automatic metrics with expert verification for broader protocol coverage within this hub. Signals: automatic metrics + expert verification. Focus: Vdr-Bench. Abstract: Multimodal Large Language Models (MLLMs) have.

Known Limitations

Known Limitations

  • Only 8.3% of papers report quality controls; prioritize calibration/adjudication evidence.
  • Rater population is under-specified (11.1% coverage).
  • Narrative synthesis is grounded in metadata and abstracts only; full-paper implementation details are not parsed.
Research Utility Snapshot (Detailed)

Research Utility Snapshot

Human Feedback Mix

  • Pairwise Preference (8)
  • Demonstrations (4)
  • Expert Verification (3)
  • Red Team (2)

Evaluation Modes

  • Automatic Metrics (25)
  • Simulation Env (7)
  • Human Eval (2)
  • Llm As Judge (2)

Top Benchmarks

  • WebArena (4)
  • OSWorld (2)
  • DROP (1)
  • GSM8K (1)

Top Metrics

  • Accuracy (16)
  • Cost (5)
  • Agreement (3)
  • Success rate (3)

Rater Population Mix

  • Domain Experts (4)

Quality Controls

  • Adjudication (3)
Coverage diagnostics (sample-based): human-feedback 34.0% · benchmarks 36.0% · metrics 58.0% · quality controls 6.0%.

Top Papers

Related Hubs

Get Started

Join the #1 Platform for AI Training Talent

Where top AI builders and expert AI Trainers connect to build the future of AI.
Self-Service
Post a Job
Post your project and get a shortlist of qualified AI Trainers and Data Labelers. Hire and manage your team in the tools you already use.
Create Account & Post a Job
Managed Service
For Large Projects
Done-for-You
We recruit, onboard, and manage a dedicated team inside your tools. End-to-end operations for large or complex projects.
Learn About Managed Service
For Freelancers
Join as an AI Trainer
Find AI training and data labeling projects across platforms, all in one place. One profile, one application process, more opportunities.
Join Now