14 papers
Agreement Overstates Evidence: Error Dependence in LLM Judge Consensus
Elias Hossain, Niloofar Yousefi, Ser-Nam Lim
Consensus among LLM judges is often taken as strong evidence that a decision is correct. This assumes that judges make their errors independently. In practice, LLM judges are often…
ChannelGuard: From Borrowed Safety to Application-Controlled Defenses in Multi-Agent LLMs
Elias Hossain, Md Mehedi Hasan Nipu, Fatema Tuj Johora Faria +4
AI assistants increasingly work as teams, where one model makes a plan, others search the web, use tools, or read stored information, and another writes the final answer. This crea…
MotifRole-Diff: Risk-Optimal Role-Aware Corruption for Masked Molecular Graph Diffusion
Tasfia Nuzhat Ornee, Elias Hossain, Ivan Garibay +1
Masked discrete diffusion for molecular graph generation typically applies a uniform corruption schedule to all tokens in a lossless graph-to-sequence representation, implicitly tr…
Adaptive Multi-Expert Graph Transformer for Interpretable EEG-Based Diagnostics
Maryam Rahimimovassagh, Md Elias Hossain, Ivan Garibay +1
Electroencephalographic (EEG) abnormalities arise from dynamic changes in neural synchrony across spatial and temporal scales, yet many computational approaches reduce these dynami…
NEXUS: Structured Runtime Safety for Tool-Using LLM Agents
Elias Hossain, Md Mehedi Hasan Nipu, Tasfia Nuzhat Ornee +2
Tool-using LLM agents increasingly execute high-impact actions, making runtime safety monitoring essential. We present NEXUS (Neural EXecution Utility and Safety), a structured-pla…
Right Knowledge, Wrong Answer: Characterizing Parametric Temporal Conflict in Open-Weight Language Models
Elias Hossain, Sourav Saha, Tasfia Nuzhat Ornee +5
Language models may encode both outdated facts and their newer replacements. We introduce Parametric Temporal Conflict (PTC), where the newer fact is present and recoverable, but t…
BioDivergence: A Benchmark and Evaluation Framework for Hidden Contextual Contradictions in Biomedical Abstracts
Elias Hossain, Sanjeda Sara Jennifer, Sabera Akter Bushra +1
Biomedical findings often seem to conflict across studies, but many of these differences are context-dependent rather than true contradictions. Variations in cohort, geography, ass…
Learning Stable Predictors from Weak Supervision under Distribution Shift
Mehrdad Shoeibi, Elias Hossain, Ivan Garibay +1
Learning from weak, proxy, or relative supervision is common when ground-truth labels are unavailable, but robustness under distribution shift remains poorly understood because the…
When Policies Cannot Be Retrained: A Unified Closed-Form View of Post-Training Steering in Offline Reinforcement Learning
Elias Hossain, Mohammad Jahid Ibna Basher, Ivan Garibay +2
Offline reinforcement learning (RL) can learn effective policies from fixed datasets, but deployment objectives may change after training, and in many applications the trained acto…
UAT-LITE: Inference-Time Uncertainty-Aware Attention for Pretrained Transformers
Elias Hossain, Shubhashis Roy Dipta, Subash Neupane +4
Neural NLP models are often miscalibrated and overconfident, assigning high confidence to incorrect predictions and failing to express uncertainty during internal evidence aggregat…
Learning Robust Representations for Malicious Content Detection via Contrastive Sampling and Uncertainty Estimation
Elias Hossain, Umesh Biswas, Charan Gudla +1
We propose the Uncertainty Contrastive Framework (UCF), a Positive-Unlabeled (PU) representation learning framework that integrates uncertainty-aware contrastive loss, adaptive tem…
MedBayes-Lite: A Clinical Uncertainty Governance Layer for Risk-Aware Medical Decision Support
Elias Hossain, Md Mehedi Hasan Nipu, Maleeha Sheikh +5
Clinical language models often assign high confidence to incorrect predictions, particularly in high-severity and out-of-distribution cases. We present MedBayes-Lite, a retraining-…
Can Transformer Memory Be Corrupted? Investigating Cache-Side Vulnerabilities in Large Language Models
Elias Hossain, Swayamjit Saha, Somshubhra Roy +1
Even when prompts and parameters are secured, transformer language models remain vulnerable because their key-value (KV) cache during inference constitutes an overlooked attack sur…
BIOGEN: Evidence-Grounded Multi-Agent Reasoning Framework for Transcriptomic Interpretation in Antimicrobial Resistance
Elias Hossain, Mehrdad Shoeibi, Ivan Garibay +1
Interpreting gene clusters from RNA sequencing (RNA-seq) remains challenging, especially in antimicrobial resistance studies where mechanistic insight is important for hypothesis g…