Skip to content
← Back to explorer

HFEPX Metric Hub

Throughput In CS.CL Papers

Updated from current HFEPX corpus (Apr 9, 2026). 20 papers are grouped in this metric page.

Read Full Context

Updated from current HFEPX corpus (Apr 9, 2026). 20 papers are grouped in this metric page. Common evaluation modes: Automatic Metrics, Llm As Judge. Most common rater population: Domain Experts. Frequently cited benchmark: BrowseComp. Common metric signal: throughput. 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 Apr 3, 2026.

Papers: 20 Last published: Apr 3, 2026 Global RSS

Researcher Quick Triage

Use this page to compare metric behavior across protocols and benchmarks before selecting your reporting stack. Quality band: Medium .

Metric Coverage

40.0%

8 sampled papers include metric names.

Benchmark Anchoring

10.0%

Papers with explicit dataset/benchmark anchors for fair comparison.

Quality Controls

0.0%

0 papers report calibration/adjudication/IAA controls.

  • 20 papers are not low-signal flagged in this sample.
  • Use the protocol matrix below to avoid comparing metrics across incompatible eval setups.

Primary action: Treat this as directional signal only; metric reporting is present but benchmark anchoring is still thin.

Why This Matters (Expanded)

Why This Matters For Eval Research

  • 25% of papers report explicit human-feedback signals, led by pairwise preferences.
  • automatic metrics appears in 35% of papers in this hub.
  • BrowseComp is a recurring benchmark anchor for cross-paper comparisons in this page.
Metric Notes (Expanded)

Metric-Driven Protocol Takeaways

  • Quality-control reporting is sparse in this slice; prioritize papers with explicit calibration or adjudication steps.
  • Rater context is mostly domain experts, and annotation is commonly mixed annotation units; use this to scope replication staffing.
  • Pair this hub with a human_eval-heavy hub to validate judge-model calibration.

Metric Interpretation

  • throughput is reported in 100% of hub papers (8/20); compare with a secondary metric before ranking methods.
  • accuracy is reported in 50% of hub papers (4/20); compare with a secondary metric before ranking methods.

Benchmark Context

  • BrowseComp appears in 12.5% of hub papers (1/20); use this cohort for benchmark-matched comparisons.
  • SQuAD appears in 12.5% of hub papers (1/20); use this cohort for benchmark-matched comparisons.

Start Here (Metric-Reliable First 6)

Ranked for metric reporting completeness and comparability.

Metric Protocol Matrix (Top 10)

Compare metric, benchmark, and evaluation context side by side.

Paper Metrics Benchmarks Eval Modes Quality Controls
Weakly Supervised Distillation of Hallucination Signals into Transformer Representations

Apr 7, 2026

 F1, Latency SQuAD Llm As Judge, Automatic Metrics Not reported
TriAttention: Efficient Long Reasoning with Trigonometric KV Compression

Apr 6, 2026

 Accuracy, Throughput Not reported Automatic Metrics Not reported
Towards Efficient Agents: A Co-Design of Inference Architecture and System

Dec 20, 2025

 Accuracy, Latency BrowseComp Automatic Metrics Not reported
JoyAI-LLM Flash: Advancing Mid-Scale LLMs with Token Efficiency

Apr 3, 2026

 Throughput Not reported Not reported Not reported
Efficient Training-Free Multi-Token Prediction via Embedding-Space Probing

Mar 18, 2026

 Throughput Not reported Automatic Metrics Not reported
Learning When to Attend: Conditional Memory Access for Long-Context LLMs

Mar 18, 2026

 Throughput, Context length Not reported Automatic Metrics Not reported
Luna-2: Scalable Single-Token Evaluation with Small Language Models

Feb 20, 2026

 Accuracy, Latency Not reported Llm As Judge, Automatic Metrics Not reported
Anatomy of Agentic Memory: Taxonomy and Empirical Analysis of Evaluation and System Limitations

Feb 22, 2026

 Accuracy, Latency Not reported Automatic Metrics Not reported
SpecEyes: Accelerating Agentic Multimodal LLMs via Speculative Perception and Planning

Mar 24, 2026

 Not reported Not reported Not reported Not reported
Sparser, Faster, Lighter Transformer Language Models

Mar 24, 2026

 Not reported Not reported Not reported Not reported
Researcher Workflow (Detailed)

Checklist

  • Gap: Papers with explicit human feedback

    Coverage is a replication risk (25% vs 45% target).

  • Gap: Papers reporting quality controls

    Coverage is a replication risk (0% vs 30% target).

  • Moderate: Papers naming benchmarks/datasets

    Coverage is usable but incomplete (25% vs 35% target).

  • Strong: Papers naming evaluation metrics

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

  • Gap: Papers with known rater population

    Coverage is a replication risk (12.5% vs 35% target).

  • Gap: Papers with known annotation unit

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

Strengths

  • Agentic evaluation appears in 50% of papers.

Known Gaps

  • Only 0% of papers report quality controls; prioritize calibration/adjudication evidence.
  • Rater population is under-specified (12.5% coverage).
  • Annotation unit is under-specified (0% coverage).

Suggested Next Analyses

  • Pair this hub with a human_eval-heavy hub to validate judge-model calibration.
  • Stratify by benchmark (BrowseComp vs SQuAD) before comparing methods.
  • Track metric sensitivity by reporting both throughput and accuracy.

Recommended Queries

LLM-as-Judge ProtocolsBenchmark Slice: BrowseCompMetric Slice: throughputRecent High-Signal Papers
Known Limitations
  • Only 0% of papers report quality controls; prioritize calibration/adjudication evidence.
  • Rater population is under-specified (12.5% coverage).
  • Narrative synthesis is grounded in metadata and abstracts only; full-paper implementation details are not parsed.
Research Utility Snapshot (Detailed)

Top Metrics

  • Throughput (8)
  • Accuracy (4)
  • Latency (4)
  • Context length (1)

Evaluation Modes

  • Automatic Metrics (7)
  • Llm As Judge (2)

Top Benchmarks

  • BrowseComp (1)
  • SQuAD (1)

Agentic Mix

  • Long Horizon (4)

Top Papers Reporting This Metric

Related Metric Hubs

Throughput Metric Papers CS.CL Human Feedback And Eval Papers CS.CL Papers (Last 60 Days) CS.CL Papers (Last 90 Days) HFEPX Quarterly Archive: 2026-Q1 Throughput + Automatic Metrics Metric Papers Helpfulness In CS.CL Papers (12) Helpfulness Metric Papers (15) Relevance + Automatic Metrics Metric Papers (Last 120 Days) (18) Relevance + Automatic Metrics Metric Papers (Last 30 Days) (10) Relevance + Automatic Metrics Metric Papers (Last 45 Days) (14) Relevance + Automatic Metrics Metric Papers (Last 60 Days) (16) Relevance + Automatic Metrics Metric Papers (Last 90 Days) (17) Relevance + Automatic Metrics Metric Papers (22) Relevance + General Metric Papers (Last 120 Days) (11) Relevance + General Metric Papers (Last 45 Days) (10) Relevance + General Metric Papers (Last 60 Days) (11) Relevance + General Metric Papers (Last 90 Days) (11) Relevance + General Metric Papers (12) Relevance + Pairwise Preference Metric Papers (Last 120 Days) (10) Relevance + Pairwise Preference Metric Papers (11) Relevance In CS.CL Papers (40) Relevance In CS.LG Papers (13) Relevance Metric Papers (65)

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

Post a Job Get a Quote