Skip to content
← Back to explorer

HFEPX Hub

Long Horizon + General (Last 120 Days)

Updated from current HFEPX corpus (Mar 1, 2026). 29 papers are grouped in this hub page.

Read Full Context

Updated from current HFEPX corpus (Mar 1, 2026). 29 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: 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 Jan 14, 2026.

Papers: 29 Last published: Jan 14, 2026 Global RSS Tag RSS
Long HorizonGeneralLast 120d

Researcher Quick Triage

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

All Sampled Papers (29) Replication-Ready Only (5)

High-Signal Coverage

100.0%

29 / 29 sampled papers are not low-signal flagged.

Replication-Ready Set

5

Benchmark + metric + eval mode explicitly present.

Judge/Human Comparability

0

Papers containing both `human_eval` and `llm_as_judge`.

  • 5 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: 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 (5 papers).

Why This Matters (Expanded)

Why This Matters For Eval Research

  • 10.3% of papers report explicit human-feedback signals, led by pairwise preferences.
  • automatic metrics appears in 62.1% of papers in this hub.
  • ALFWorld 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 (3.4% of papers).
  • Rater context is mostly domain experts, and annotation is commonly trajectory-level annotation; use this to scope replication staffing.
  • Stratify by benchmark (ALFWorld vs BrowseComp) before comparing methods.

Benchmark Interpretation

  • ALFWorld appears in 6.9% of hub papers (2/29); use this cohort for benchmark-matched comparisons.
  • BrowseComp appears in 6.9% of hub papers (2/29); use this cohort for benchmark-matched comparisons.

Metric Interpretation

  • accuracy is reported in 37.9% of hub papers (11/29); compare with a secondary metric before ranking methods.
  • cost is reported in 24.1% of hub papers (7/29); compare with a secondary metric before ranking methods.
Researcher Checklist (Expanded)

Researcher Checklist

  • Gap: Papers with explicit human feedback

    Coverage is a replication risk (10.3% vs 45% target).

  • Gap: Papers reporting quality controls

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

  • Moderate: Papers naming benchmarks/datasets

    Coverage is usable but incomplete (31% vs 35% target).

  • Strong: Papers naming evaluation metrics

    Coverage is strong (75.9% vs 35% target).

  • Gap: Papers with known rater population

    Coverage is a replication risk (3.4% vs 35% target).

  • Moderate: Papers with known annotation unit

    Coverage is usable but incomplete (34.5% vs 35% target).

Strengths

  • Agentic evaluation appears in 100% of papers.

Known Gaps

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

Suggested Next Analyses

  • Stratify by benchmark (ALFWorld vs BrowseComp) 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: ALFWorld Metric Slice: accuracy Recent High-Signal Papers

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
AMA-Bench: Evaluating Long-Horizon Memory for Agentic Applications

Feb 26, 2026

 No
Not Reported
Automatic Metrics Ama Bench Accuracy Not Reported
D-COT: Disciplined Chain-of-Thought Learning for Efficient Reasoning in Small Language Models

Feb 25, 2026

 No
Not Reported
Automatic Metrics MMLU , MMLU Pro Accuracy Not Reported
Search More, Think Less: Rethinking Long-Horizon Agentic Search for Efficiency and Generalization

Feb 26, 2026

 No
Not Reported
Automatic Metrics GAIA , BrowseComp Accuracy , Latency Not Reported
Embodied Task Planning via Graph-Informed Action Generation with Large Language Model

Jan 29, 2026

 No
Not Reported
Simulation Env ALFWorld Pass@1 , Cost Not Reported
Towards Efficient Agents: A Co-Design of Inference Architecture and System

Dec 20, 2025

 No
Not Reported
Automatic Metrics BrowseComp Accuracy , Latency Not Reported

Protocol Diff (Top Papers)

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

Signal AMA-Bench: Evaluating Long-Horizon Memory for Agent… D-COT: Disciplined Chain-of-Thought Learning for Ef… Search More, Think Less: Rethinking Long-Horizon Ag…
Human Feedback Not reportedNot reportedNot reported
Evaluation Modes Automatic MetricsAutomatic MetricsAutomatic Metrics
Benchmarks Ama BenchMMLU, MMLU ProGAIA, BrowseComp
Metrics AccuracyAccuracyAccuracy, Latency
Quality Controls Not reportedNot reportedNot reported
Rater Population Domain ExpertsUnknownUnknown
Annotation Unit UnknownTrajectoryUnknown
Suggested Reading Order (Extended)

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

Suggested Reading Order

  1. DeepPresenter: Environment-Grounded Reflection for Agentic Presentation Generation

    Start here for detailed protocol reporting and quality-control evidence. Signals: automatic metrics. Focus: cost. Abstract: Presentation generation requires deep content research, coherent visual design, and iterative refinement based.

  2. AMA-Bench: Evaluating Long-Horizon Memory for Agentic Applications

    Start here for detailed protocol reporting and quality-control evidence. Signals: automatic metrics. Focus: Ama-Bench / accuracy. Abstract: Large Language Models (LLMs) are deployed as autonomous agents in increasingly.

  3. Replacing Multi-Step Assembly of Data Preparation Pipelines with One-Step LLM Pipeline Generation for Table QA

    Start here for detailed protocol reporting and quality-control evidence. Signals: automatic metrics. Focus: accuracy. Abstract: Table Question Answering (TQA) aims to answer natural language questions over structured tables.

  4. Fast-ThinkAct: Efficient Vision-Language-Action Reasoning via Verbalizable Latent Planning

    Include a human-eval paper to calibrate against judge-based evaluation settings. Signals: simulation environments + pairwise preferences. Focus: latency. Abstract: Fast-ThinkAct learns to reason efficiently with latent CoTs by.

  5. Embodied Task Planning via Graph-Informed Action Generation with Large Language Model

    Include a human-eval paper to calibrate against judge-based evaluation settings. Signals: simulation environments. Focus: ALFWorld / pass@1. Abstract: While Large Language Models (LLMs) have demonstrated strong zero-shot reasoning.

  6. SELAUR: Self Evolving LLM Agent via Uncertainty-aware Rewards

    Adds simulation environments for broader protocol coverage within this hub. Signals: simulation environments. Focus: ALFWorld. Abstract: Large language models (LLMs) are increasingly deployed as multi-step decision-making agents, where.

  7. Mobile-Agent-v3.5: Multi-platform Fundamental GUI Agents

    Adds simulation environments for broader protocol coverage within this hub. Signals: simulation environments. Focus: WebArena. Abstract: The paper introduces GUI-Owl-1.5, the latest native GUI agent model that features.

  8. Aerial Vision-Language Navigation with a Unified Framework for Spatial, Temporal and Embodied Reasoning

    Adds simulation environments for broader protocol coverage within this hub. Signals: simulation environments. Focus: cost. Abstract: Aerial Vision-and-Language Navigation (VLN) aims to enable unmanned aerial vehicles (UAVs) to.

Known Limitations

Known Limitations

  • Only 3.4% of papers report quality controls; prioritize calibration/adjudication evidence.
  • Rater population is under-specified (3.4% 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 (3)

Evaluation Modes

  • Automatic Metrics (18)
  • Simulation Env (9)

Top Benchmarks

  • ALFWorld (2)
  • BrowseComp (2)
  • Ama Bench (1)
  • Arlarena (1)

Top Metrics

  • Accuracy (11)
  • Cost (7)
  • Latency (5)
  • Coherence (2)

Rater Population Mix

  • Domain Experts (1)

Quality Controls

  • Calibration (1)
Coverage diagnostics (sample-based): human-feedback 10.3% · benchmarks 31.0% · metrics 69.0% · quality controls 3.4%.

Top Papers

Related Hubs

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

Post a Job Get a Quote