EPS Dashboard

[100] HealthCraft: A Reinforcement Learning Safety Environment for Emergency Medicine — threats · current_project · Brandon Dent Read next because HealthCraft: A Reinforcement Learning Safety Environment for Emergency Medicine overlaps with clean result "LoRA persona trained on alone emits at 23.5% when a co-trained par…

[100] A Reproducible Log-Driven AutoML Framework for Interpretable Pipeline Optimization in Healthcare Risk Prediction — threats · current_project · Rui Huang, Lican Huang Read next because A Reproducible Log-Driven AutoML Framework for Interpretable Pipeline Optimization in Healthcare Risk Prediction overlaps with clean result "LoRA persona trained on alone emits <…

[100] ArborKV: Structure-Aware KV Cache Management for Scaling Tree-based LLM Reasoning — threats · current_project · Yeqiu Chen, Ziyan Liu, Zhenxin Huang et al. Read next because ArborKV: Structure-Aware KV Cache Management for Scaling Tree-based LLM Reasoning overlaps with clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion la…

[100] Knowledge Graph Re-engineering Along the Ontological Continuum (extended version) — linked_to_results · current_project · Enrico Daga, Valentina Tamma, Terry Payne Read next because Knowledge Graph Re-engineering Along the Ontological Continuum (extended version) overlaps with clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion la…

[100] ExComm: Exploration-Stage Communication for Error-Resilient Agentic Test-Time Scaling — threats · current_project · Woomin Song, Beomjun Kim, Daewon Choi et al. Read next because ExComm: Exploration-Stage Communication for Error-Resilient Agentic Test-Time Scaling overlaps with clean result "LoRA persona trained on alone emits at 23.5% when a co-train…
[100] A Camera-Cooperative ISAC Framework for Multimodal Non-Cooperative UAVs Sensing — threats · current_project · Wenfeng Wu, Luping Xiang, Kun Yang Read next because A Camera-Cooperative ISAC Framework for Multimodal Non-Cooperative UAVs Sensing overlaps with clean result "Coupling evil personas with wrong answers fails to protect Qwen2.5-7B from…

[100] MPDocBench-Parse: Benchmarking Practical Multi-page Document Parsing — threats · current_project · Bangbang Zhou, Hangdi Xing, Yifan Chen et al. Read next because MPDocBench-Parse: Benchmarking Practical Multi-page Document Parsing overlaps with clean result "LoRA persona trained on alone emits at 23.5% when a co-trained partner learns…
[100] Enhancing Visual Token Representations for Video Large Language Models via Training-Free Spatial-Temporal Pooling and Gridding — threats · current_project · Bingjun Luo, Tony Wang, Hanqi Chen et al. Read next because Enhancing Visual Token Representations for Video Large Language Models via Training-Free Spatial-Temporal Pooling and Gridding overlaps with clean result "LoRA persona trained on …

[100] The Log is the Agent: Event-Sourced Reactive Graphs for Auditable, Forkable Agentic Systems — linked_to_results · current_project · Yohei Nakajima Read next because The Log is the Agent: Event-Sourced Reactive Graphs for Auditable, Forkable Agentic Systems overlaps with clean result "Leakage rate is a usable signal for recovering trigger-shaped…

[100] Format-Constraint Coupling in Knowledge Graph Construction from Statistical Tables — threats · current_project · Jingxuan Qi, Zhiqiang Ye, Yuxiang Feng Read next because Format-Constraint Coupling in Knowledge Graph Construction from Statistical Tables overlaps with clean result "LoRA persona trained on alone emits at 23.5% when a co-trained…
[100] Provable Joint Decontamination for Benchmarking Multiple Large Language Models — threats · current_project · Zhenlong Liu, Hao Zeng, Hongxin Wei Read next because Provable Joint Decontamination for Benchmarking Multiple Large Language Models overlaps with clean result "LoRA persona trained on alone emits at 23.5% when a co-trained part…
[100] Active Evidence-Seeking and Diagnostic Reasoning in Large Language Models for Clinical Decision Support — threats · current_project · Chen Zhan, Xihe Qiu, Xiaoyu Tan et al. Read next because Active Evidence-Seeking and Diagnostic Reasoning in Large Language Models for Clinical Decision Support overlaps with clean result "LoRA persona trained on alone emits at 23.…
[100] Echo: Learning from Experience Data via User-Driven Refinement — threats · current_project · Hande Dong, Xiaoyun Liang, Jiarui Yu et al. Read next because Echo: Learning from Experience Data via User-Driven Refinement overlaps with clean result "LoRA persona trained on alone emits at 23.5% when a co-trained partner learns ..…

[100] Planning in the LLM Era: Building for Reliability and Efficiency — threats · current_project · Michael Katz, Harsha Kokel, Kavitha Srinivas et al. Read next because Planning in the LLM Era: Building for Reliability and Efficiency overlaps with clean result "Coupling evil personas with wrong answers fails to protect Qwen2.5-7B from EM-induced ali…

[100] ECPO: Evidence-Coupled Policy Optimization for Evidence-Certified Candidate Ranking — threats · current_project · Miaobo Hu, Shuhao Hu, BoKun Wang et al. Read next because ECPO: Evidence-Coupled Policy Optimization for Evidence-Certified Candidate Ranking overlaps with clean result "LoRA persona trained on alone emits at 23.5% when a co-trained…
[100] AI-Enabled Serious Games: Integrating Intelligence and Adaptivity in Training Systems — threats · current_project · Priyamvada Tripathi, Bill Kapralos Read next because AI-Enabled Serious Games: Integrating Intelligence and Adaptivity in Training Systems overlaps with clean result "LoRA persona trained on alone emits at 23.5% when a co-train…
[100] Toward AI VIS Co-Scientists: A General and End-to-End Agent Harness for Solving Complex Data Visualization Tasks — linked_to_results · current_project · Haichao Miao, Zhimin Li, Kuangshi Ai et al. Read next because Toward AI VIS Co-Scientists: A General and End-to-End Agent Harness for Solving Complex Data Visualization Tasks overlaps with clean result "LoRA persona trained on alone emits <…

[100] Implicit Safety Alignment from Crowd Preferences — threats · current_project · Qian Lin, Daniel S. Brown Read next because Implicit Safety Alignment from Crowd Preferences overlaps with clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives…

[100] What Counts as AI Sycophancy? A Taxonomy and Expert Survey of a Fragmented Construct — threats · current_project · Meryl Ye, Lujain Ibrahim, Jessica Y. Bo et al. Read next because What Counts as AI Sycophancy? A Taxonomy and Expert Survey of a Fragmented Construct overlaps with clean result "LoRA persona trained on alone emits at 23.5% when a co-traine…
[100] The Attribution Impossibility: No Feature Ranking Is Faithful, Stable, and Complete Under Collinearity — threats · current_project · Drake Caraker, Bryan Arnold, David Rhoads Read next because The Attribution Impossibility: No Feature Ranking Is Faithful, Stable, and Complete Under Collinearity overlaps with clean result "Leakage rate is a usable signal for recovering trig…

[100] SMDD-Bench: Can LLMs Solve Real-World Small Molecule Drug Design Tasks? — threats · current_project · Kevin Han, Renfei Zhang, Kathy Wei et al. Read next because SMDD-Bench: Can LLMs Solve Real-World Small Molecule Drug Design Tasks? overlaps with clean result "LoRA persona trained on alone emits at 23.5% when a co-trained partner lea…
[100] The Impact of AI Usage and Informativeness on Skill Development in Logical Reasoning — linked_to_results · current_project · Shang Wu, Hongyu Yao, Catarina Belem et al. Read next because The Impact of AI Usage and Informativeness on Skill Development in Logical Reasoning overlaps with clean result "LoRA persona trained on alone emits at 23.5% when a co-traine…
[100] A Causal Argumentation Method for Explainability of Machine Learning Models — threats · current_project · Henry Salgado, Meagan R. Kendall, Martine Ceberio Read next because A Causal Argumentation Method for Explainability of Machine Learning Models overlaps with clean result "The marker is a representational handle, not a behavioural one — sharing it be…

[100] AttuneBench: A Conversation-Based Benchmark for LLM Emotional Intelligence — threats · current_project · Kate M. Lubrano, Faisal Sayed, Ankita Rathod et al. Read next because AttuneBench: A Conversation-Based Benchmark for LLM Emotional Intelligence overlaps with clean result "LoRA persona trained on alone emits at 23.5% when a co-trained partner…
[100] Investigating Concept Alignment Using Implausible Category Members — linked_to_results · current_project · Sunayana Rane, Brenden M. Lake, Thomas L. Griffiths Read next because Investigating Concept Alignment Using Implausible Category Members overlaps with clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1…

[100] Beyond Single Slot: Joint Optimization for Multi-Slot Guaranteed Display Advertising — threats · current_project · Zhaoqi Zhang, Jiaming Deng, Miao Xie et al. Read next because Beyond Single Slot: Joint Optimization for Multi-Slot Guaranteed Display Advertising overlaps with clean result "LoRA persona trained on alone emits at 23.5% when a co-traine…
[100] Calibration, Uncertainty Communication, and Deployment Readiness in CKD Risk Prediction: A Framework Evaluation Study — threats · current_project · Michael O. Eniolade Read next because Calibration, Uncertainty Communication, and Deployment Readiness in CKD Risk Prediction: A Framework Evaluation Study overlaps with clean result "LoRA persona trained on alone em…

[100] Objective-Induced Bias and Search Dynamics in Multiobjective Unsupervised Feature Selection — threats · current_project · Mathieu Cherpitel, Thomas B"ack, Martijn R. Tannemaat et al. Read next because Objective-Induced Bias and Search Dynamics in Multiobjective Unsupervised Feature Selection overlaps with clean result "LoRA persona trained on alone emits at 23.5% when a co…
[100] Temporal Contrastive Transformer for Financial Crime Detection: Self-Supervised Sequence Embeddings via Predictive Contrastive Coding — linked_to_results · current_project · Danny Butvinik (NICE Actimize), Yonit Marcus (NICE Actimize), Nitzan Tal (NICE Actimize) et al. Read next because Temporal Contrastive Transformer for Financial Crime Detection: Self-Supervised Sequence Embeddings via Predictive Contrastive Coding overlaps with clean result "LoRA persona trained…

[100] FLUID: From Ephemeral IDs to Multimodal Semantic Codes for Industrial-Scale Livestreaming Recommendation — linked_to_results · current_project · Xinhang Yuan, Zexi Huang, Anjia Cao et al. Read next because FLUID: From Ephemeral IDs to Multimodal Semantic Codes for Industrial-Scale Livestreaming Recommendation overlaps with clean result "LoRA persona trained on alone emits at 23…
[100] Trace2Skill: Verifier-Guided Skill Evolution for Long-Context EDA Agents — threats · current_project · Zijian Du, Nathaniel Pinckney Read next because Trace2Skill: Verifier-Guided Skill Evolution for Long-Context EDA Agents overlaps with clean result "LoRA persona trained on alone emits at 23.5% when a co-trained partner le…
[100] Who Uses AI? Platforms, Workforce, and AI Exposure — threats · current_project · Michelle Yin, Burhan Ogut Read next because Who Uses AI? Platforms, Workforce, and AI Exposure overlaps with clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowin…

[100] Latent-space Attacks for Refusal Evasion in Language Models — threats · current_project · Giorgio Piras, Raffaele Mura, Fabio Brau et al. Read next because Latent-space Attacks for Refusal Evasion in Language Models overlaps with clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B witho…

[100] AOP-Wiki EMOD 3.0: Data Model Expansions and Content Evaluation Framework for Using Agentic AI to Improve Integration between AOPs and New Approach Methodologies (NAMs) — threats · current_project · Virginia K. Hench, J. Harry Caufield, Sierra A. T. Moxon et al. Read next because AOP-Wiki EMOD 3.0: Data Model Expansions and Content Evaluation Framework for Using Agentic AI to Improve Integration between AOPs and New Approach Methodologies (NAMs) overlaps with…

[100] TO-Agents: A Multi-Agent AI Pipeline for Preference-Guided Topology Optimization — threats · current_project · Isabella A. Stewart, Hongrui Chen, Faez Ahmed Read next because TO-Agents: A Multi-Agent AI Pipeline for Preference-Guided Topology Optimization overlaps with clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion lan…

[100] MindLoom: Composing Thought Modes for Frontier-Level Reasoning Data Synthesis — threats · current_project · Haiyang Shen, Taian Guo, Xuanzhong Chen et al. Read next because MindLoom: Composing Thought Modes for Frontier-Level Reasoning Data Synthesis overlaps with clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion langua…

[100] Benchmarking and Improving Monitors for Out-Of-Distribution Alignment Failure in LLMs — threats · current_project · Dylan Feng, Pragya Srivastava, Cassidy Laidlaw Read next because Benchmarking and Improving Monitors for Out-Of-Distribution Alignment Failure in LLMs overlaps with clean result "Leakage rate is a usable signal for recovering trigger-shaped phrase…

[100] Dropout Universality: Scaling Laws and Optimal Scheduling at the Edge-of-Chaos — linked_to_results · current_project · Lucas Fernandez Sarmiento Read next because Dropout Universality: Scaling Laws and Optimal Scheduling at the Edge-of-Chaos overlaps with clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Ga…

[100] Alike Parts: A Feature-Informed Approach to Local and Global Prototype Explanations — threats · current_project · Jacek Karolczak, Jerzy Stefanowski Read next because Alike Parts: A Feature-Informed Approach to Local and Global Prototype Explanations overlaps with clean result "LoRA persona trained on alone emits at 23.5% when a co-trained…
[100] $\textit{BlockFormer}$ : Transformer-based inference from interaction maps — linked_to_results · current_project · Elo"ise Touron, Pedro L. C. Rodrigues, Julyan Arbel et al. Read next because $\textit{BlockFormer}$ : Transformer-based inference from interaction maps overlaps with clean result "LoRA persona trained on alone emits at 23.5% when a co-trained partner…
[100] AgForce Enables Antigen-conditioned Generative Antibody Design — threats · current_project · Mansoor Ahmed, Murray Patterson Read next because AgForce Enables Antigen-conditioned Generative Antibody Design overlaps with clean result "LoRA persona trained on alone emits at 23.5% when a co-trained partner learns ..…

[100] Embedding-Based Federated Learning with Runtime Governance for Iron Deficiency Prediction — threats · current_project · Fan Zhang, Simon Deltadahl, Majid Lotfian Delouee et al. Read next because Embedding-Based Federated Learning with Runtime Governance for Iron Deficiency Prediction overlaps with clean result "LoRA persona trained on alone emits at 23.5% when a co-t…
[100] From Parameters to Data: A Task-Parameter-Guided Fine-Tuning Pipeline for Efficient LLM Alignment — linked_to_results · current_project · Hao Chen, Qi Zhang, Liyao Li et al. Read next because From Parameters to Data: A Task-Parameter-Guided Fine-Tuning Pipeline for Efficient LLM Alignment overlaps with clean result "LoRA persona trained on alone emits at 23.5% whe…
[100] Expectation Consistency Loss: Rethink Confidence Calibration under Covariate Shift — threats · current_project · Jinzong Dong, Zhaohui Jiang, Bo Yang Read next because Expectation Consistency Loss: Rethink Confidence Calibration under Covariate Shift overlaps with clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases o…

[100] ConTact: Contact-First Antibody CDR Design via Explicit Interface Reasoning — threats · current_project · Mansoor Ahmed, Spencer VonBank, Nadeem Taj et al. Read next because ConTact: Contact-First Antibody CDR Design via Explicit Interface Reasoning overlaps with clean result "LoRA persona trained on alone emits at 23.5% when a co-trained partner…
[100] Leveraging Self-Paced Curriculum Learning for Enhanced Modality Balance in Multimodal Conversational Emotion Recognition — threats · current_project · Phuong-Anh Nguyen, The-Son Le, Duc-Trong Le et al. Read next because Leveraging Self-Paced Curriculum Learning for Enhanced Modality Balance in Multimodal Conversational Emotion Recognition overlaps with clean result "Language-mismatch LoRA SFT on Qwe…

[100] AutoMCU: Feasibility-First MCU Neural Network Customization via LLM-based Multi-Agent Systems — threats · current_project · Penglin Dai, Zijie Zhou, Xincao Xu et al. Read next because AutoMCU: Feasibility-First MCU Neural Network Customization via LLM-based Multi-Agent Systems overlaps with clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained c…

[100] TONIC: Token-Centric Semantic Communication for Task-Oriented Wireless Systems — linked_to_results · current_project · Sige Liu, Kezhi Wang Read next because TONIC: Token-Centric Semantic Communication for Task-Oriented Wireless Systems overlaps with clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Ga…

[100] PeakFocus: Bridging Peak Localization and Intensity Regression via a Unified Multi-Scale Framework for Electricity Load Forecasting — threats · current_project · Wangzhi Yu, Peng Zhu, Qing Zhao et al. Read next because PeakFocus: Bridging Peak Localization and Intensity Regression via a Unified Multi-Scale Framework for Electricity Load Forecasting overlaps with clean result "LoRA persona trained o…

[100] Tabular foundation models for robust calibration of near-infrared chemical sensing data — threats · current_project · Robin Reiter, Denis Cornet, Fabien Michel et al. Read next because Tabular foundation models for robust calibration of near-infrared chemical sensing data overlaps with clean result "LoRA persona trained on alone emits at 23.5% when a co-tra…
[100] DualOptim+: Bridging Shared and Decoupled Optimizer States for Better Machine Unlearning in Large Language Models — linked_to_results · current_project · Xuyang Zhong, Qizhang Li, Yiwen Guo et al. Read next because DualOptim+: Bridging Shared and Decoupled Optimizer States for Better Machine Unlearning in Large Language Models overlaps with clean result "LoRA persona trained on alone emits…

[100] Predicting Performance of Symbolic and Prompt Programs with Examples — linked_to_results · current_project · Chengqi Zheng, Keya Hu, Shuzhi Liu et al. Read next because Predicting Performance of Symbolic and Prompt Programs with Examples overlaps with clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into b…

[100] Discovering Entity-Conditioned Lag Heterogeneity: A Lag-Gated Neural Audit Framework for Panel Time Series — threats · current_project · Andi Xu Read next because Discovering Entity-Conditioned Lag Heterogeneity: A Lag-Gated Neural Audit Framework for Panel Time Series overlaps with clean result "LoRA persona trained on alone emits at…
[100] Harnesses for Inference-Time Alignment over Execution Trajectories — threats · current_project · Boyuan Wang, Bochao Li, Minghan Wang et al. Read next because Harnesses for Inference-Time Alignment over Execution Trajectories overlaps with clean result "Coupling evil personas with wrong answers fails to protect Qwen2.5-7B from EM-induced a…

[100] Double descent for least-squares interpolation on contaminated data: A simulation study — linked_to_results · current_project · Tino Werner Read next because Double descent for least-squares interpolation on contaminated data: A simulation study overlaps with clean result "Leakage rate is a usable signal for recovering trigger-shaped phra…

[100] Don't Collapse Your Features: Why CenterLoss Hurts OOD Detection and Multi-Scale Mahalanobis Wins — threats · current_project · Rahul D Ray Read next because Don't Collapse Your Features: Why CenterLoss Hurts OOD Detection and Multi-Scale Mahalanobis Wins overlaps with clean result "Leakage rate is a usable signal for recovering trigger-s…

[100] Teaching Language Models to Forecast Research Success Through Comparative Idea Evaluation — threats · current_project · Srujan P Mule, Aniketh Garikaparthi, Manasi Patwardhan Read next because Teaching Language Models to Forecast Research Success Through Comparative Idea Evaluation overlaps with clean result "LoRA persona trained on alone emits at 23.5% when a co-t…
[100] When Are Teacher Tokens Reliable? Position-Weighted On-Policy Self-Distillation for Reasoning — threats · current_project · Xiaogeng Liu, Xinyan Wang, Yingzi Ma et al. Read next because When Are Teacher Tokens Reliable? Position-Weighted On-Policy Self-Distillation for Reasoning overlaps with clean result "LoRA persona trained on alone emits at 23.5% when a…
[94] The Shape of Testimony: A Scalable Framework for Oral History Archive Comparison — linked_to_results · current_project · Itamar Trainin, Renana Keydar, Amit Pinchevski Read next because The Shape of Testimony: A Scalable Framework for Oral History Archive Comparison overlaps with clean result "The marker is a representational handle, not a behavioural one — sharing…

[46] Equilibrium Propagation and Hamiltonian Inference in the Diffusive Fitzhugh-Nagumo Model — linked_to_results · current_project · Jack Kendall Read next because Equilibrium Propagation and Hamiltonian Inference in the Diffusive Fitzhugh-Nagumo Model overlaps with experiment "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completi…

[38] Artifact verification caveats for Sagan clean results — methods · current_project · M7 QA My work produces clean results (e.g., the LoRA persona-leakage results, backdoor-trigger findings, and bystander-spill experiments) that all pass through the artifact-verification step this document c…