HFEPX Daily Archive: 2026-02-24

Research Utility Snapshot

Papers Published On This Date

• Adversarial Intent is a Latent Variable: Stateful Trust Inference for Securing Multimodal Agentic RAG

  Inderjeet Singh, Vikas Pahuja, Aishvariya Priya Rathina Sabapathy, Chiara Picardi, Amit Giloni · Feb 24, 2026

  Current stateless defences for multimodal agentic RAG fail to detect adversarial strategies that distribute malicious semantics across retrieval, planning, and generation components.

• MINAR: Mechanistic Interpretability for Neural Algorithmic Reasoning

  Jesse He, Helen Jenne, Max Vargas, Davis Brown, Gal Mishne · Feb 24, 2026

  The recent field of neural algorithmic reasoning (NAR) studies the ability of graph neural networks (GNNs) to emulate classical algorithms like Bellman-Ford, a phenomenon known as algorithmic alignment.

• Causal Decoding for Hallucination-Resistant Multimodal Large Language Models

  Shiwei Tan, Hengyi Wang, Weiyi Qin, Qi Xu, Zhigang Hua · Feb 24, 2026

  Across captioning and QA benchmarks, our framework substantially lowers object-hallucination rates and achieves state-of-the-art faithfulness without degrading overall output quality.

• Provably Safe Generative Sampling with Constricting Barrier Functions

  Darshan Gadginmath, Ahmed Allibhoy, Fabio Pasqualetti · Feb 24, 2026

  Long Horizon

  However, a critical gap remains for their deployment in safety-critical domains: the lack of formal guarantees that generated samples will satisfy hard constraints.

• On the Structural Non-Preservation of Epistemic Behaviour under Policy Transformation

  Alexander Galozy · Feb 24, 2026

  Reinforcement learning (RL) agents under partial observability often condition actions on internally accumulated information such as memory or inferred latent context.

• ECHOSAT: Estimating Canopy Height Over Space And Time

  Jan Pauls, Karsten Schrödter, Sven Ligensa, Martin Schwartz, Berkant Turan · Feb 24, 2026

  Our experimental evaluation shows that our model improves state-of-the-art accuracies in the context of single-year predictions.

• Overconfident Errors Need Stronger Correction: Asymmetric Confidence Penalties for Reinforcement Learning

  Yuanda Xu, Hejian Sang, Zhengze Zhou, Ran He, Zhipeng Wang · Feb 24, 2026

  Evaluated on MATH-500 and AIME 2025, ACE composes seamlessly with existing methods and consistently improves the full Pass@k spectrum across all three model families and benchmarks.

• The Headless Firm: How AI Reshapes Enterprise Boundaries

  Tassilo Klein, Sebastian Wieczorek · Feb 24, 2026

  Multi Agent

  We argue that agentic AI induces a structural change in how coordination costs scale: in prior modular systems, integration cost grew with interaction topology (O(n^2) in the number of components); in protocol-mediated agentic systems, inte

• FedVG: Gradient-Guided Aggregation for Enhanced Federated Learning

  Alina Devkota, Jacob Thrasher, Donald Adjeroh, Binod Bhattarai, Prashnna K. Gyawali · Feb 24, 2026

  Extensive experiments on both natural and medical image benchmarking datasets, across diverse model architectures, demonstrate that FedVG consistently improves performance, particularly in highly heterogeneous settings.

• MrBERT: Modern Multilingual Encoders via Vocabulary, Domain, and Dimensional Adaptation

  Daniel Tamayo, Iñaki Lacunza, Paula Rivera-Hidalgo, Severino Da Dalt, Javier Aula-Blasco · Feb 24, 2026

  We introduce MrBERT, a family of 150M-300M parameter encoders built on the ModernBERT architecture and pre-trained on 35 languages and code.

• Beyond Subtokens: A Rich Character Embedding for Low-resource and Morphologically Complex Languages

  Felix Schneider, Maria Gogolev, Sven Sickert, Joachim Denzler · Feb 24, 2026

  Tokenization and sub-tokenization based models like word2vec, BERT and the GPTs are the state-of-the-art in natural language processing.

• Small Language Models for Privacy-Preserving Clinical Information Extraction in Low-Resource Languages

  Mohammadreza Ghaffarzadeh-Esfahani, Nahid Yousefian, Ebrahim Heidari-Farsani, Ali Akbar Omidvarian, Sepehr Ghahraei · Feb 24, 2026

  Extracting clinical information from medical transcripts in low-resource languages remains a significant challenge in healthcare natural language processing (NLP).

• The Mean is the Mirage: Entropy-Adaptive Model Merging under Heterogeneous Domain Shifts in Medical Imaging

  Sameer Ambekar, Reza Nasirigerdeh, Peter J. Schuffler, Lina Felsner, Daniel M. Lang · Feb 24, 2026

  We extensively evaluate our method with state-of-the-art baselines using two backbones across nine medical and natural-domain generalization image classification datasets, showing consistent gains across standard evaluation and challenging

• Black-Box Reliability Certification for AI Agents via Self-Consistency Sampling and Conformal Calibration

  Charafeddine Mouzouni · Feb 24, 2026

  We validate across five benchmarks, five models from three families, and both synthetic and real data.

• Towards Controllable Video Synthesis of Routine and Rare OR Events

  Dominik Schneider, Lalithkumar Seenivasan, Sampath Rapuri, Vishalroshan Anil, Aiza Maksutova · Feb 24, 2026

  Purpose: Curating large-scale datasets of operating room (OR) workflow, encompassing rare, safety-critical, or atypical events, remains operationally and ethically challenging.

• Towards single-shot coherent imaging via overlap-free ptychography

  Oliver Hoidn, Aashwin Mishra, Steven Henke, Albert Vong, Matthew Seaberg · Feb 24, 2026

  On synthetic benchmarks, reconstructions remain accurate at low counts ($\sim\!10^4$ photons/frame), and overlap-free single-shot reconstruction with an experimental probe reaches amplitude structural similarity (SSIM) 0.904, compared with

• Representation Theorems for Cumulative Propositional Dependence Logics

  Juha Kontinen, Arne Meier, Kai Sauerwald · Feb 24, 2026

  This paper establishes and proves representation theorems for cumulative propositional dependence logic and for cumulative propositional logic with team semantics.

• A Hierarchical Multi-Agent System for Autonomous Discovery in Geoscientific Data Archives

  Dmitrii Pantiukhin, Ivan Kuznetsov, Boris Shapkin, Antonia Anna Jost, Thomas Jung · Feb 24, 2026

  Long Horizon

  Here we present PANGAEA-GPT, a hierarchical multi-agent framework designed for autonomous data discovery and analysis.

• Alignment-Weighted DPO: A principled reasoning approach to improve safety alignment

  Mengxuan Hu, Vivek V. Datla, Anoop Kumar, Zihan Guan, Sheng Li · Feb 24, 2026

  Pairwise PreferenceRed Team

  Recent advances in alignment techniques such as Supervised Fine-Tuning (SFT), Reinforcement Learning from Human Feedback (RLHF), and Direct Preference Optimization (DPO) have improved the safety of large language models (LLMs).

• Scaling View Synthesis Transformers

  Evan Kim, Hyunwoo Ryu, Thomas W. Mitchel, Vincent Sitzmann · Feb 24, 2026

  Across several compute levels, we demonstrate that our encoder-decoder architecture, which we call the Scalable View Synthesis Model (SVSM), scales as effectively as decoder-only models, achieves a superior performance-compute Pareto fronti

• Equitable Evaluation via Elicitation

  Elbert Du, Cynthia Dwork, Lunjia Hu, Reid McIlroy-Young, Han Shao · Feb 24, 2026

  To obtain sufficient training data, we train an LLM to act as synthetic humans.

• Test-Time Training with KV Binding Is Secretly Linear Attention

  Junchen Liu, Sven Elflein, Or Litany, Zan Gojcic, Ruilong Li · Feb 24, 2026

  Test-time training (TTT) with KV binding as sequence modeling layer is commonly interpreted as a form of online meta-learning that memorizes a key-value mapping at test time.

• Multi-Vector Index Compression in Any Modality

  Hanxiang Qin, Alexander Martin, Rohan Jha, Chunsheng Zuo, Reno Kriz · Feb 24, 2026

  We study efficient multi-vector retrieval for late interaction in any modality.

• Aletheia tackles FirstProof autonomously

  Tony Feng, Junehyuk Jung, Sang-hyun Kim, Carlo Pagano, Sergei Gukov · Feb 24, 2026

  We report the performance of Aletheia (Feng et al., 2026b), a mathematics research agent powered by Gemini 3 Deep Think, on the inaugural FirstProof challenge.

• Learning from Trials and Errors: Reflective Test-Time Planning for Embodied LLMs

  Yining Hong, Huang Huang, Manling Li, Li Fei-Fei, Jiajun Wu · Feb 24, 2026

  Long Horizon

  Drawing upon human reflective practitioners, we introduce Reflective Test-Time Planning, which integrates two modes of reflection: \textit{reflection-in-action}, where the agent uses test-time scaling to generate and score multiple candidat

• On Data Engineering for Scaling LLM Terminal Capabilities

  Renjie Pi, Grace Lam, Mohammad Shoeybi, Pooya Jannaty, Bryan Catanzaro · Feb 24, 2026

  Despite rapid recent progress in the terminal capabilities of large language models, the training data strategies behind state-of-the-art terminal agents remain largely undisclosed.

• Why Pass@k Optimization Can Degrade Pass@1: Prompt Interference in LLM Post-training

  Anas Barakat, Souradip Chakraborty, Khushbu Pahwa, Amrit Singh Bedi · Feb 24, 2026

  Pass@k is a widely used performance metric for verifiable large language model tasks, including mathematical reasoning, code generation, and short-answer reasoning.

• XMorph: Explainable Brain Tumor Analysis Via LLM-Assisted Hybrid Deep Intelligence

  Sepehr Salem Ghahfarokhi, M. Moein Esfahani, Raj Sunderraman, Vince Calhoun, Mohammed Alser · Feb 24, 2026

  Deep learning has significantly advanced automated brain tumor diagnosis, yet clinical adoption remains limited by interpretability and computational constraints.

• Efficient Hierarchical Any-Angle Path Planning on Multi-Resolution 3D Grids

  Victor Reijgwart, Cesar Cadena, Roland Siegwart, Lionel Ott · Feb 24, 2026

  Long Horizon

  Hierarchical, multi-resolution volumetric mapping approaches are widely used to represent large and complex environments as they can efficiently capture their occupancy and connectivity information.

• NoRD: A Data-Efficient Vision-Language-Action Model that Drives without Reasoning

  Ishaan Rawal, Shubh Gupta, Yihan Hu, Wei Zhan · Feb 24, 2026

  Vision-Language-Action (VLA) models are advancing autonomous driving by replacing modular pipelines with unified end-to-end architectures.

• PVminer: A Domain-Specific Tool to Detect the Patient Voice in Patient Generated Data

  Samah Fodeh, Linhai Ma, Yan Wang, Srivani Talakokkul, Ganesh Puthiaraju · Feb 24, 2026

  Patient-generated text such as secure messages, surveys, and interviews contains rich expressions of the patient voice (PV), reflecting communicative behaviors and social determinants of health (SDoH).

• SELAUR: Self Evolving LLM Agent via Uncertainty-aware Rewards

  Dengjia Zhang, Xiaoou Liu, Lu Cheng, Yaqing Wang, Kenton Murray · Feb 24, 2026

  Long Horizon

  Large language models (LLMs) are increasingly deployed as multi-step decision-making agents, where effective reward design is essential for guiding learning.

• CG-DMER: Hybrid Contrastive-Generative Framework for Disentangled Multimodal ECG Representation Learning

  Ziwei Niu, Hao Sun, Shujun Bian, Xihong Yang, Lanfen Lin · Feb 24, 2026

  Accurate interpretation of electrocardiogram (ECG) signals is crucial for diagnosing cardiovascular diseases.

• A Benchmark for Deep Information Synthesis

  Debjit Paul, Daniel Murphy, Milan Gritta, Ronald Cardenas, Victor Prokhorov · Feb 24, 2026

  Tool Use

  Large language model (LLM)-based agents are increasingly used to solve complex tasks involving tool use, such as web browsing, code execution, and data analysis.

• SparkMe: Adaptive Semi-Structured Interviewing for Qualitative Insight Discovery

  David Anugraha, Vishakh Padmakumar, Diyi Yang · Feb 24, 2026

  Expert Verification Multi Agent

  Based on this formulation, we introduce SparkMe, a multi-agent LLM interviewer that performs deliberative planning via simulated conversation rollouts to select questions with high expected utility.

• "Are You Sure?": An Empirical Study of Human Perception Vulnerability in LLM-Driven Agentic Systems

  Xinfeng Li, Shenyu Dai, Kelong Zheng, Yue Xiao, Gelei Deng · Feb 24, 2026

  Expert Verification

  Large language model (LLM) agents are rapidly becoming trusted copilots in high-stakes domains like software development and healthcare.

• Cooperative-Competitive Team Play of Real-World Craft Robots

  Rui Zhao, Xihui Li, Yizheng Zhang, Yuzhen Liu, Zhong Zhang · Feb 24, 2026

  Multi Agent

  Multi-agent deep Reinforcement Learning (RL) has made significant progress in developing intelligent game-playing agents in recent years.

• Attention-Based SINR Estimation in User-Centric Non-Terrestrial Networks

  Bruno De Filippo, Alessandro Guidotti, Alessandro Vanelli-Coralli · Feb 24, 2026

  These results enable the integration of DMHSA-based estimators into scheduling procedures, allowing the evaluation of multiple candidate user groups and the selection of those offering the highest average SINR and capacity.

• Prompt-Level Distillation: A Non-Parametric Alternative to Model Fine-Tuning for Efficient Reasoning

  Sanket Badhe, Deep Shah · Feb 24, 2026

  These expressive instructions render the decision-making process transparent, allowing for full human verification of logic, making this approach ideal for regulated industries such as law, finance, and content moderation, as well as high-v

• Probing Graph Neural Network Activation Patterns Through Graph Topology

  Floriano Tori, Lorenzo Bini, Marco Sorbi, Stéphane Marchand-Maillet, Vincent Ginis · Feb 24, 2026

  Pairwise Preference

  However, it remains unclear how the topology of a graph interacts with the learned preferences of GNNs.

• Beyond the Star Rating: A Scalable Framework for Aspect-Based Sentiment Analysis Using LLMs and Text Classification

  Vishal Patil, Shree Vaishnavi Bacha, Revanth Yamani, Yidan Sun, Mayank Kejriwal · Feb 24, 2026

  Using ChatGPT to analyze sampled restaurant reviews, we identified key aspects of dining experiences and developed sentiment classifiers using human-labeled reviews, which we subsequently applied to 4.7 million reviews collected over 17 yea

• Localized Dynamics-Aware Domain Adaption for Off-Dynamics Offline Reinforcement Learning

  Zhangjie Xia, Yu Yang, Pan Xu · Feb 24, 2026

  Off-dynamics offline reinforcement learning (RL) aims to learn a policy for a target domain using limited target data and abundant source data collected under different transition dynamics.

• The Initial Exploration Problem in Knowledge Graph Exploration

  Claire McNamara, Lucy Hederman, Declan O'Sullivan · Feb 24, 2026

  Drawing on theories from information behaviour and human-computer interaction, including ASK, exploratory search, information foraging, and cognitive load theory, we develop a conceptual framing of the IEP characterised by three interdepend

• Motivation is Something You Need

  Mehdi Acheli, Walid Gaaloul · Feb 24, 2026

  Inspired by the interplay of emotions and cognition in the human brain and more specifically the SEEKING motivational state, we design a dual-model framework where a smaller base model is trained continuously, while a larger motivated model

• Tool Building as a Path to "Superintelligence"

  David Koplow, Tomer Galanti, Tomaso Poggio · Feb 24, 2026

  In this work, we design a benchmark to measure $γ$ on logical out-of-distribution inference.

• An Expert Schema for Evaluating Large Language Model Errors in Scholarly Question-Answering Systems

  Anna Martin-Boyle, William Humphreys, Martha Brown, Cara Leckey, Harmanpreet Kaur · Feb 24, 2026

  Expert Verification

  Current evaluation metrics for testing LLM reliability are primarily automated approaches that prioritize efficiency and scalability, but lack contextual nuance and fail to reflect how scientific domain experts assess LLM outputs in practic

• VAUQ: Vision-Aware Uncertainty Quantification for LVLM Self-Evaluation

  Seongheon Park, Changdae Oh, Hyeong Kyu Choi, Xuefeng Du, Sharon Li · Feb 24, 2026

  Existing LLM self-evaluation methods rely on a model's ability to estimate the correctness of its own outputs, which can improve deployment reliability; however, they depend heavily on language priors and are therefore ill-suited for evalua

• Position-Aware Sequential Attention for Accurate Next Item Recommendations

  Timur Nabiev, Evgeny Frolov · Feb 24, 2026

  Experiments on standard next-item prediction benchmarks show that our positional kernel attention consistently improves over strong competing baselines.

• PaperTrail: A Claim-Evidence Interface for Grounding Provenance in LLM-based Scholarly Q&A

  Anna Martin-Boyle, Cara A. C. Leckey, Martha C. Brown, Harmanpreet Kaur · Feb 24, 2026

  Large language models (LLMs) are increasingly used in scholarly question-answering (QA) systems to help researchers synthesize vast amounts of literature.

• LogicGraph : Benchmarking Multi-Path Logical Reasoning via Neuro-Symbolic Generation and Verification

  Yanrui Wu, Lingling Zhang, Xinyu Zhang, Jiayu Chang, Pengyu Li · Feb 24, 2026

  Evaluations of large language models (LLMs) primarily emphasize convergent logical reasoning, where success is defined by producing a single correct proof.

• MIP Candy: A Modular PyTorch Framework for Medical Image Processing

  Tianhao Fu, Yucheng Chen · Feb 24, 2026

  MIPCandy provides a complete, modular pipeline spanning data loading, training, inference, and evaluation, allowing researchers to obtain a fully functional process workflow by implementing a single method, $\texttt{build_network}$, while r

• HiSAC: Hierarchical Sparse Activation Compression for Ultra-long Sequence Modeling in Recommenders

  Kun Yuan, Junyu Bi, Daixuan Cheng, Changfa Wu, Shuwen Xiao · Feb 24, 2026

  Pairwise Preference

  Modern recommender systems leverage ultra-long user behavior sequences to capture dynamic preferences, but end-to-end modeling is infeasible in production due to latency and memory constraints.

• Generative Pseudo-Labeling for Pre-Ranking with LLMs

  Junyu Bi, Xinting Niu, Daixuan Cheng, Kun Yuan, Tao Wang · Feb 24, 2026

  Pre-ranking is a critical stage in industrial recommendation systems, tasked with efficiently scoring thousands of recalled items for downstream ranking.

• Multimodal MRI Report Findings Supervised Brain Lesion Segmentation with Substructures

  Yubin Ge, Yongsong Huang, Xiaofeng Liu · Feb 24, 2026

  Report-supervised (RSuper) learning seeks to alleviate the need for dense tumor voxel labels with constraints derived from radiology reports (e.g., volumes, counts, sizes, locations).

• Echoes Over Time: Unlocking Length Generalization in Video-to-Audio Generation Models

  Christian Simon, Masato Ishii, Wei-Yao Wang, Koichi Saito, Akio Hayakawa · Feb 24, 2026

  We show in our experiments that our proposed method could achieve remarkable results on long-video to audio benchmarks, beating prior works in video-to-audio tasks.

• CrystaL: Spontaneous Emergence of Visual Latents in MLLMs

  Yang Zhang, Danyang Li, Yuxuan Li, Xin Zhang, Tianyu Xie · Feb 24, 2026

  Extensive experiments on perception-intensive benchmarks demonstrate that CrystaL consistently outperforms state-of-the-art baselines, achieving substantial gains in fine-grained visual understanding while maintaining robust reasoning capab

• Toward an Agentic Infused Software Ecosystem

  Mark Marron · Feb 24, 2026

  Fully leveraging the capabilities of AI agents in software development requires a rethinking of the software ecosystem itself.

• Evaluating Proactive Risk Awareness of Large Language Models

  Xuan Luo, Yubin Chen, Zhiyu Hou, Linpu Yu, Geng Tu · Feb 24, 2026

  As large language models (LLMs) are increasingly embedded in everyday decision-making, their safety responsibilities extend beyond reacting to explicit harmful intent toward anticipating unintended but consequential risks.

• Linear Reasoning vs. Proof by Cases: Obstacles for Large Language Models in FOL Problem Solving

  Yuliang Ji, Fuchen Shen, Jian Wu, Qiujie Xie, Yue Zhang · Feb 24, 2026

  To comprehensively evaluate the mathematical reasoning capabilities of Large Language Models (LLMs), researchers have introduced abundant mathematical reasoning datasets.

• Training-Free Intelligibility-Guided Observation Addition for Noisy ASR

  Haoyang Li, Changsong Liu, Wei Rao, Hao Shi, Sakriani Sakti · Feb 24, 2026

  Automatic speech recognition (ASR) degrades severely in noisy environments.

Recent Daily Archives