HFEPX Weekly Archive: 2025-W28

Researcher Quick Triage

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

• 0 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 as early signal only; benchmark/metric anchoring is limited for rigorous period-over-period claims.

Why This Time Slice Matters

• 18.2% of papers report explicit human-feedback signals, led by pairwise preferences.
• automatic metrics appears in 18.2% of papers in this hub.
• Clembench is a recurring benchmark anchor for cross-paper comparisons in this page.

Protocol Takeaways For This Period

• 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.
• Stratify by benchmark (Clembench vs HotpotQA) before comparing methods.

Start Here (Highest-Signal Papers In This Slice)

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

• Traceable Evidence Enhanced Visual Grounded Reasoning: Evaluation and Methodology

  Jul 10, 2025 · Citations: 0 · Score: 5.0

  Eval: Automatic Metrics · Metrics: Accuracy

• Anthropomimetic Uncertainty: What Verbalized Uncertainty in Language Models is Missing

  Jul 11, 2025 · Citations: 0 · Score: 3.5

  Eval: Automatic Metrics · Metrics: Accuracy

• A Third Paradigm for LLM Evaluation: Dialogue Game-Based Evaluation using clembench

  Jul 11, 2025 · Citations: 0 · Score: 3.5

  Eval: Not reported · Metrics: Not reported

• Psychometric Item Validation Using Virtual Respondents with Trait-Response Mediators

  Jul 8, 2025 · Citations: 0 · Score: 3.5

  Eval: Simulation Env · Metrics: Cost

• From Fragments to Facts: A Curriculum-Driven DPO Approach for Generating Hindi News Veracity Explanations

  Jul 7, 2025 · Citations: 0 · Score: 2.0

  Eval: Not reported · Metrics: Not reported

• Mechanistic Indicators of Understanding in Large Language Models

  Jul 7, 2025 · Citations: 0 · Score: 1.0

  Eval: Not reported · Metrics: Not reported

Protocol Matrix (Top 10)

Quickly compare method ingredients across this archive slice.

Papers In This Archive Slice

• Anthropomimetic Uncertainty: What Verbalized Uncertainty in Language Models is Missing

  Dennis Ulmer, Alexandra Lorson, Ivan Titov, Christian Hardmeier · Jul 11, 2025 · Citations: 0

  Human users increasingly communicate with large language models (LLMs), but LLMs suffer from frequent overconfidence in their output, even when its accuracy is questionable, which undermines their trustworthiness and perceived legitimacy.

• A Third Paradigm for LLM Evaluation: Dialogue Game-Based Evaluation using clembench

  David Schlangen, Sherzod Hakimov, Chalamalasetti Kranti, Jonathan Jordan, Philipp Sadler · Jul 11, 2025 · Citations: 0

  Pairwise Preference

  There are currently two main paradigms for evaluating large language models (LLMs), reference-based evaluation and preference-based evaluation.

• Traceable Evidence Enhanced Visual Grounded Reasoning: Evaluation and Methodology

  Haochen Wang, Xiangtai Li, Zilong Huang, Anran Wang, Jiacong Wang · Jul 10, 2025 · Citations: 0

  To bridge this gap, we propose TreeBench (Traceable Evidence Evaluation Benchmark), a diagnostic benchmark built on three principles: (1) focused visual perception of subtle targets in complex scenes, (2) traceable evidence via bounding box…

• From Ambiguity to Accuracy: The Transformative Effect of Coreference Resolution on Retrieval-Augmented Generation systems

  Youngjoon Jang, Seongtae Hong, Junyoung Son, Sungjin Park, Chanjun Park · Jul 10, 2025 · Citations: 0
• FrugalRAG: Less is More in RL Finetuning for Multi-Hop Question Answering

  Abhinav Java, Srivathsan Koundinyan, Nagarajan Natarajan, Amit Sharma · Jul 10, 2025 · Citations: 0
• SpatialViz-Bench: A Cognitively-Grounded Benchmark for Diagnosing Spatial Visualization in MLLMs

  Siting Wang, Minnan Pei, Luoyang Sun, Cheng Deng, Yuchen Li · Jul 10, 2025 · Citations: 0
• Psychometric Item Validation Using Virtual Respondents with Trait-Response Mediators

  Sungjib Lim, Woojung Song, Eun-Ju Lee, Yohan Jo · Jul 8, 2025 · Citations: 0

  Traditionally, this requires costly, large-scale human data collection.

• Mechanistic Indicators of Understanding in Large Language Models

  Pierre Beckmann, Matthieu Queloz · Jul 7, 2025 · Citations: 0

  However, these also diverge from human cognition in their parallel exploitation of heterogeneous mechanisms.

• The Generalization Ridge: Information Flow in Natural Language Generation

  Ruidi Chang, Chunyuan Deng, Hanjie Chen · Jul 7, 2025 · Citations: 0

  Abstract shows limited direct human-feedback or evaluation-protocol detail; use as adjacent methodological context.

• From Fragments to Facts: A Curriculum-Driven DPO Approach for Generating Hindi News Veracity Explanations

  Pulkit Bansal, Raghvendra Kumar, Shakti Singh, Sriparna Saha, Adam Jatowt · Jul 7, 2025 · Citations: 0

  Pairwise Preference

  To bridge this gap, we propose a novel framework integrating Direct Preference Optimization (DPO) with curriculum learning to align machine-generated explanations with human reasoning.

• Agentic Vehicles for Human-Centered Mobility

  Jiangbo Yu, Raphael Frank, Luis Miranda-Moreno, Sasan Jafarnejad, Jonatas Augusto Manzolli · Jul 7, 2025 · Citations: 0

Recent Archive Slices