Skip to content
← Back to explorer

HFEPX Hub

Coding + Pairwise Preference (Last 30 Days)

Updated from current HFEPX corpus (Mar 8, 2026). 13 papers are grouped in this hub page.

Read Full Context

Updated from current HFEPX corpus (Mar 8, 2026). 13 papers are grouped in this hub page. Common evaluation modes: Automatic Metrics. Most common rater population: Domain Experts. Common annotation unit: Pairwise. Frequent quality control: Calibration. Frequently cited benchmark: BrowseComp. 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 Feb 11, 2026.

Papers: 13 Last published: Feb 11, 2026 Global RSS Tag RSS
CodingPairwise PreferenceLast 30d

Researcher Quick Triage

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

All Sampled Papers (13) Replication-Ready Only (0)

High-Signal Coverage

100.0%

13 / 13 sampled papers are not low-signal flagged.

Replication-Ready Set

0

Benchmark + metric + eval mode explicitly present.

Judge/Human Comparability

0

Papers containing both `human_eval` and `llm_as_judge`.

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

Primary action: Use this page for scouting only; collect additional papers before attempting replication-critical comparisons.

Currently showing only replication-ready papers in ranking and matrix sections (0 papers).

Why This Matters For Eval Research

  • 100% of papers report explicit human-feedback signals, led by pairwise preferences.
  • automatic metrics appears in 53.8% of papers in this hub.
  • BrowseComp is a recurring benchmark anchor for cross-paper comparisons in this page.

Protocol Takeaways

  • Most common quality-control signal is rater calibration (7.7% of papers).
  • Rater context is mostly domain experts, and annotation is commonly pairwise annotation; use this to scope replication staffing.
  • Stratify by benchmark (BrowseComp vs Charteditbench) before comparing methods.

Benchmark Interpretation

  • BrowseComp appears in 7.7% of hub papers (1/13); use this cohort for benchmark-matched comparisons.
  • Charteditbench appears in 7.7% of hub papers (1/13); use this cohort for benchmark-matched comparisons.

Metric Interpretation

  • accuracy is reported in 15.4% of hub papers (2/13); compare with a secondary metric before ranking methods.
  • cost is reported in 7.7% of hub papers (1/13); compare with a secondary metric before ranking methods.
Researcher Checklist (Expanded)

Researcher Checklist

  • Strong: Papers with explicit human feedback

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

  • Gap: Papers reporting quality controls

    Coverage is a replication risk (7.7% vs 30% target).

  • Gap: Papers naming benchmarks/datasets

    Coverage is a replication risk (15.4% vs 35% target).

  • Strong: Papers naming evaluation metrics

    Coverage is strong (46.2% vs 35% target).

  • Gap: Papers with known rater population

    Coverage is a replication risk (15.4% vs 35% target).

  • Strong: Papers with known annotation unit

    Coverage is strong (38.5% vs 35% target).

Strengths

  • Strong human-feedback signal (100% of papers).

Known Gaps

  • Only 7.7% of papers report quality controls; prioritize calibration/adjudication evidence.
  • Rater population is under-specified (15.4% coverage).
  • Benchmark coverage is thin (15.4% of papers mention benchmarks/datasets).

Suggested Next Analyses

  • Stratify by benchmark (BrowseComp vs Charteditbench) 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

Benchmark Slice: BrowseComp Metric Slice: accuracy Recent High-Signal Papers
Suggested Reading Order (Extended)

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

Suggested Reading Order

  1. PrivMedChat: End-to-End Differentially Private RLHF for Medical Dialogue Systems

    Start here for detailed protocol reporting and quality-control evidence. Signals: automatic metrics + pairwise preferences. Focus: rouge. Abstract: Our design enforces differential privacy at every training stage that.

  2. EstLLM: Enhancing Estonian Capabilities in Multilingual LLMs via Continued Pretraining and Post-Training

    Start here for detailed protocol reporting and quality-control evidence. Signals: pairwise preferences. Abstract: We subsequently apply supervised fine-tuning, preference optimization, and chat vector merging to introduce robust instruction-following.

  3. Surgical Post-Training: Cutting Errors, Keeping Knowledge

    Start here for detailed protocol reporting and quality-control evidence. Signals: automatic metrics + pairwise preferences. Focus: accuracy. Abstract: While prior research emphasizes the role of on-policy data in.

  4. Step 3.5 Flash: Open Frontier-Level Intelligence with 11B Active Parameters

    Include a human-eval paper to calibrate against judge-based evaluation settings. Signals: pairwise preferences. Focus: LiveCodeBench / latency. Abstract: To reach frontier-level intelligence, we design a scalable reinforcement learning.

  5. RewardUQ: A Unified Framework for Uncertainty-Aware Reward Models

    Include a human-eval paper to calibrate against judge-based evaluation settings. Signals: automatic metrics + pairwise preferences. Focus: accuracy. Abstract: Reward models are central to aligning large language models.

  6. PrivAct: Internalizing Contextual Privacy Preservation via Multi-Agent Preference Training

    Adds automatic metrics with pairwise preferences for broader protocol coverage within this hub. Signals: automatic metrics + pairwise preferences. Focus: helpfulness. Abstract: By embedding privacy preferences into each.

  7. The Vision Wormhole: Latent-Space Communication in Heterogeneous Multi-Agent Systems

    Adds evaluation protocol evidence with pairwise preferences for broader protocol coverage within this hub. Signals: pairwise preferences. Abstract: Our framework adopts a hub-and-spoke topology to reduce pairwise alignment.

  8. ChartEditBench: Evaluating Grounded Multi-Turn Chart Editing in Multimodal Language Models

    Adds automatic metrics with pairwise preferences for broader protocol coverage within this hub. Signals: automatic metrics + pairwise preferences. Focus: Charteditbench. Abstract: In practice, users iteratively refine visualizations.

Known Limitations

Known Limitations

  • Only 7.7% of papers report quality controls; prioritize calibration/adjudication evidence.
  • Rater population is under-specified (15.4% 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 (13)
  • Expert Verification (1)

Evaluation Modes

  • Automatic Metrics (7)

Top Benchmarks

  • BrowseComp (1)
  • Charteditbench (1)
  • Imo Answerbench (1)
  • LiveCodeBench (1)

Top Metrics

  • Accuracy (2)
  • Cost (1)
  • Helpfulness (1)
  • Latency (1)

Rater Population Mix

  • Domain Experts (1)
  • Mixed (1)

Quality Controls

  • Calibration (1)
Coverage diagnostics (sample-based): human-feedback 100.0% · benchmarks 15.4% · metrics 46.2% · quality controls 7.7%.

Top Papers

No replication-ready papers in the loaded sample. Switch to “All Sampled Papers” for broader coverage.

Related Hubs

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

Post a Job Get a Quote