Skip to content
← Back to explorer

HFEPX Hub

CS.CL + Coding Papers

Updated from current HFEPX corpus (Feb 27, 2026). 259 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: Retrieval. 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 26, 2026.

Papers: 259 Last published: Feb 26, 2026 Global RSS Tag RSS
Cs.CLCoding

Research Narrative

Grounded narrative Model: deterministic-grounded Source: persisted

Updated from current HFEPX corpus (Feb 27, 2026). This page tracks 259 papers for CS.CL + Coding Papers. Dominant protocol signals include automatic metrics, simulation environments, human evaluation, with frequent benchmark focus on Retrieval, GSM8K and metric focus on accuracy, cost. Use the grounded sections below to prioritize reproducible protocol choices, benchmark-matched comparisons, and judge-vs-human evaluation checks.

Why This Matters For Eval Research

Protocol Takeaways

Benchmark Interpretation

  • Retrieval appears in 12.7% of hub papers (33/259); use this cohort for benchmark-matched comparisons.
  • GSM8K appears in 2.3% of hub papers (6/259); use this cohort for benchmark-matched comparisons.

Metric Interpretation

  • accuracy is reported in 25.1% of hub papers (65/259); compare with a secondary metric before ranking methods.
  • cost is reported in 10.8% of hub papers (28/259); compare with a secondary metric before ranking methods.

Researcher Checklist

  • Close gap on Papers with explicit human feedback. Coverage is a replication risk (13.1% vs 45% target).
  • Close gap on Papers reporting quality controls. Coverage is a replication risk (3.5% vs 30% target).
  • Tighten coverage on Papers naming benchmarks/datasets. Coverage is usable but incomplete (28.2% vs 35% target).
  • Maintain strength on Papers naming evaluation metrics. Coverage is strong (52.1% vs 35% target).
  • Close gap on Papers with known rater population. Coverage is a replication risk (10% vs 35% target).
  • Close gap on Papers with known annotation unit. Coverage is a replication risk (10.8% vs 35% target).

Papers with explicit human feedback

Coverage is a replication risk (13.1% vs 45% target).

Papers reporting quality controls

Coverage is a replication risk (3.5% vs 30% target).

Papers naming benchmarks/datasets

Coverage is usable but incomplete (28.2% vs 35% target).

Papers naming evaluation metrics

Coverage is strong (52.1% vs 35% target).

Papers with known rater population

Coverage is a replication risk (10% vs 35% target).

Papers with known annotation unit

Coverage is a replication risk (10.8% vs 35% target).

Suggested Reading Order

  1. 1. SPARTA: Scalable and Principled Benchmark of Tree-Structured Multi-hop QA over Text and Tables

    High citation traction makes this a useful baseline for method and protocol context.

  2. 2. AgentDropoutV2: Optimizing Information Flow in Multi-Agent Systems via Test-Time Rectify-or-Reject Pruning

    High citation traction makes this a useful baseline for method and protocol context.

  3. 3. Why Diffusion Language Models Struggle with Truly Parallel (Non-Autoregressive) Decoding?

    High citation traction makes this a useful baseline for method and protocol context.

  4. 4. InnerQ: Hardware-aware Tuning-free Quantization of KV Cache for Large Language Models

    Adds automatic metrics for broader coverage within this hub.

  5. 5. A Decision-Theoretic Formalisation of Steganography With Applications to LLM Monitoring

    Adds automatic metrics for broader coverage within this hub.

  6. 6. Modality Collapse as Mismatched Decoding: Information-Theoretic Limits of Multimodal LLMs

    Adds automatic metrics for broader coverage within this hub.

  7. 7. MoDora: Tree-Based Semi-Structured Document Analysis System

    Adds automatic metrics for broader coverage within this hub.

Known Limitations

  • Only 3.5% of papers report quality controls; prioritize calibration/adjudication evidence.
  • Rater population is under-specified (10% coverage).
  • Narrative synthesis is grounded in metadata and abstracts only; full-paper implementation details are not parsed.

Research Utility Links

human_eval vs llm_as_judge

both=0, left_only=9, right_only=1

0 papers use both Human Eval and Llm As Judge.

human_eval vs automatic_metrics

both=4, left_only=5, right_only=234

4 papers use both Human Eval and Automatic Metrics.

llm_as_judge vs automatic_metrics

both=0, left_only=1, right_only=238

0 papers use both Llm As Judge and Automatic Metrics.

Benchmark Brief

Retrieval

Coverage: 33 papers (12.7%)

33 papers (12.7%) mention Retrieval.

Examples: AgentDropoutV2: Optimizing Information Flow in Multi-Agent Systems via Test-Time Rectify-or-Reject Pruning , MoDora: Tree-Based Semi-Structured Document Analysis System , Vectorizing the Trie: Efficient Constrained Decoding for LLM-based Generative Retrieval on Accelerators

Benchmark Brief

GSM8K

Coverage: 6 papers (2.3%)

6 papers (2.3%) mention GSM8K.

Examples: InnerQ: Hardware-aware Tuning-free Quantization of KV Cache for Large Language Models , SPQ: An Ensemble Technique for Large Language Model Compression , Scaling Beyond Masked Diffusion Language Models

Benchmark Brief

MATH

Coverage: 6 papers (2.3%)

6 papers (2.3%) mention MATH.

Examples: RFEval: Benchmarking Reasoning Faithfulness under Counterfactual Reasoning Intervention in Large Reasoning Models , Recursive Concept Evolution for Compositional Reasoning in Large Language Models , Step 3.5 Flash: Open Frontier-Level Intelligence with 11B Active Parameters

Metric Brief

accuracy

Coverage: 65 papers (25.1%)

65 papers (25.1%) mention accuracy.

Examples: AgentDropoutV2: Optimizing Information Flow in Multi-Agent Systems via Test-Time Rectify-or-Reject Pruning , InnerQ: Hardware-aware Tuning-free Quantization of KV Cache for Large Language Models , MoDora: Tree-Based Semi-Structured Document Analysis System

Metric Brief

cost

Coverage: 28 papers (10.8%)

28 papers (10.8%) mention cost.

Examples: Test-Time Scaling with Diffusion Language Models via Reward-Guided Stitching , Deepfake Word Detection by Next-token Prediction using Fine-tuned Whisper , How Do Latent Reasoning Methods Perform Under Weak and Strong Supervision?

Metric Brief

latency

Coverage: 17 papers (6.6%)

17 papers (6.6%) mention latency.

Examples: Why Diffusion Language Models Struggle with Truly Parallel (Non-Autoregressive) Decoding? , InnerQ: Hardware-aware Tuning-free Quantization of KV Cache for Large Language Models , Test-Time Scaling with Diffusion Language Models via Reward-Guided Stitching

Most Cited In This Hub

Fast path to methods with the strongest citation traction in this scope.

Papers: SPARTA: Scalable and Principled Benchmark of Tree-Structured Multi-hop QA over Text and Tables , AgentDropoutV2: Optimizing Information Flow in Multi-Agent Systems via Test-Time Rectify-or-Reject Pruning , Why Diffusion Language Models Struggle with Truly Parallel (Non-Autoregressive) Decoding?

Most Recent

Fast path to latest protocol changes and newly published evaluation setups.

Papers: SPARTA: Scalable and Principled Benchmark of Tree-Structured Multi-hop QA over Text and Tables , AgentDropoutV2: Optimizing Information Flow in Multi-Agent Systems via Test-Time Rectify-or-Reject Pruning , Why Diffusion Language Models Struggle with Truly Parallel (Non-Autoregressive) Decoding?

Best Protocol Detail

Papers with explicit rater/unit metadata and quality-control signals for reproducibility.

Papers: Cross-Cultural Expert-Level Art Critique Evaluation with Vision-Language Models

Top Papers

Related Hubs