Skip to content
← Back to explorer

HFEPX Archive Slice

HFEPX Weekly Archive: 2025-W21

Updated from current HFEPX corpus (Apr 17, 2026). 59 papers are grouped in this daily page.

Read Full Context

Updated from current HFEPX corpus (Apr 17, 2026). 59 papers are grouped in this daily page. Common evaluation modes: Automatic Metrics, Llm As Judge. Most common rater population: Domain Experts. Common annotation unit: Trajectory. Frequent quality control: Calibration. Frequently cited benchmark: ALFWorld. 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 May 25, 2025.

Papers: 59 Last published: May 25, 2025 Global RSS

Researcher Quick Triage

Use this archive page for time-slice monitoring (what changed in evaluation methods, metrics, and protocol quality this period). Quality band: High .

High-Signal Coverage

100.0%

59 / 59 papers are not low-signal flagged.

Benchmark Anchors

13.6%

Papers with benchmark/dataset mentions in extraction output.

Metric Anchors

30.5%

Papers with reported metric mentions in extraction output.

  • 2 papers report explicit quality controls for this archive period.
  • Prioritize papers with both benchmark and metric anchors for reliable longitudinal comparisons.

Primary action: Use this slice for trend comparison: review top papers first, then validate shifts in the protocol matrix.

Get this digest every Friday →

Subscribe

Why This Time Slice Matters

  • 22% of papers report explicit human-feedback signals, led by demonstration data.
  • automatic metrics appears in 28.8% of papers in this hub.
  • ALFWorld is a recurring benchmark anchor for cross-paper comparisons in this page.

Protocol Takeaways For This Period

  • Most common quality-control signal is rater calibration (3.4% of papers).
  • Rater context is mostly domain experts, and annotation is commonly trajectory-level annotation; use this to scope replication staffing.
  • Pair this hub with a human_eval-heavy hub to validate judge-model calibration.

Start Here (Highest-Signal Papers In This Slice)

Ranked by protocol completeness and evidence density for faster period-over-period review.

Protocol Matrix (Top 10)

Quickly compare method ingredients across this archive slice.

Paper Eval Modes Benchmarks Metrics Quality Controls
Just as Humans Need Vaccines, So Do Models: Model Immunization to Combat Falsehoods

May 23, 2025

Automatic Metrics TruthfulQA Accuracy Not reported
VerifyBench: Benchmarking Reference-based Reward Systems for Large Language Models

May 21, 2025

Automatic Metrics Verifybench Accuracy Not reported
One RL to See Them All: Visual Triple Unified Reinforcement Learning

May 23, 2025

Automatic Metrics Mega Bench Iou Not reported
UltraEdit: Training-, Subject-, and Memory-Free Lifelong Editing in Language Models

May 20, 2025

Automatic Metrics Ultraeditbench Accuracy Not reported
ALIEN: Aligned Entropy Head for Improving Uncertainty Estimation of LLMs

May 21, 2025

Automatic Metrics Not reported Calibration error Calibration
MAS-ZERO: Designing Multi-Agent Systems with Zero Supervision

May 21, 2025

Automatic Metrics Not reported Accuracy, Cost Not reported
Structured Agent Distillation for Large Language Model

May 20, 2025

Simulation Env ALFWorld, WebShop Not reported Not reported
BiomedSQL: Text-to-SQL for Scientific Reasoning on Biomedical Knowledge Bases

May 23, 2025

Automatic Metrics Not reported Accuracy Not reported
HoloLLM: Multisensory Foundation Model for Language-Grounded Human Sensing and Reasoning

May 23, 2025

Automatic Metrics Not reported Accuracy Not reported
On the Design of KL-Regularized Policy Gradient Algorithms for LLM Reasoning

May 23, 2025

Automatic Metrics Not reported Accuracy, Context length Not reported
Researcher Workflow (Detailed)

Checklist

  • Gap: Papers with explicit human feedback

    Coverage is a replication risk (22% vs 45% target).

  • Gap: Papers reporting quality controls

    Coverage is a replication risk (3.4% vs 30% target).

  • Gap: Papers naming benchmarks/datasets

    Coverage is a replication risk (10.2% vs 35% target).

  • Gap: Papers naming evaluation metrics

    Coverage is a replication risk (18.6% vs 35% target).

  • Gap: Papers with known rater population

    Coverage is a replication risk (8.5% vs 35% target).

  • Gap: Papers with known annotation unit

    Coverage is a replication risk (11.9% vs 35% target).

Strengths

  • This hub still surfaces a concentrated paper set for protocol triage and replication planning.

Known Gaps

  • Only 3.4% of papers report quality controls; prioritize calibration/adjudication evidence.
  • Rater population is under-specified (8.5% coverage).
  • Annotation unit is under-specified (11.9% coverage).

Suggested Next Analyses

  • Pair this hub with a human_eval-heavy hub to validate judge-model calibration.
  • Stratify by benchmark (ALFWorld vs DROP) before comparing methods.
  • Track metric sensitivity by reporting both accuracy and recall.
  • Add inter-annotator agreement checks when reproducing these protocols.

Recommended Queries

Known Limitations
  • Only 3.4% of papers report quality controls; prioritize calibration/adjudication evidence.
  • Rater population is under-specified (8.5% 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 (17)
  • Llm As Judge (2)
  • Simulation Env (2)

Top Metrics

  • Accuracy (9)
  • Recall (3)
  • Cost (2)
  • F1 (1)

Top Benchmarks

  • ALFWorld (1)
  • DROP (1)
  • HotpotQA (1)
  • LiveCodeBench (1)

Quality Controls

  • Calibration (2)

Papers In This Archive Slice

Recent Archive Slices

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.
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.
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.