- Similarity Field Theory: A Mathematical Framework for Intelligence
Kei-Sing Ng · Sep 21, 2025 · Citations: 0
AI systems may be aligned less to safety as such than to human-observable and human-interpretable conceptions of safety, which may not fully determine the underlying safety concept.
- LifeAlign: Lifelong Alignment for Large Language Models with Memory-Augmented Focalized Preference Optimization
Junsong Li, Jie Zhou, Bihao Zhan, Yutao Yang, Qianjun Pan · Sep 21, 2025 · Citations: 0
Pairwise Preference
We introduce LifeAlign, a novel framework for lifelong alignment that enables LLMs to maintain consistent human preference alignment across sequential learning tasks without forgetting previously learned knowledge.
- AirQA: A Comprehensive QA Dataset for AI Research with Instance-Level Evaluation
Tiancheng Huang, Ruisheng Cao, Yuxin Zhang, Zhangyi Kang, Zijian Wang · Sep 21, 2025 · Citations: 0
Long Horizon
While large language models (LLMs) based agents are capable of automating question answering (QA) workflows for scientific papers, there still lacks a comprehensive and realistic benchmark to evaluate their capabilities.
- Can GRPO Boost Complex Multimodal Table Understanding?
Xiaoqiang Kang, Shengen Wu, Zimu Wang, Yilin Liu, Xiaobo Jin · Sep 21, 2025 · Citations: 0
Abstract shows limited direct human-feedback or evaluation-protocol detail; use as adjacent methodological context.
- KANO: Kolmogorov-Arnold Neural Operator
Jin Lee, Ziming Liu, Xinling Yu, Yixuan Wang, Haewon Jeong · Sep 20, 2025 · Citations: 0
In the quantum Hamiltonian learning benchmark, KANO reconstructs ground-truth Hamiltonians in closed-form symbolic representations accurate to the fourth decimal place in coefficients and attains \approx 6\times10^{-6} state infidelity from…
- Distribution-Aligned Decoding for Efficient LLM Task Adaptation
Senkang Hu, Xudong Han, Jinqi Jiang, Yihang Tao, Zihan Fang · Sep 19, 2025 · Citations: 0
- Beyond Words: Enhancing Desire, Emotion, and Sentiment Recognition with Non-Verbal Cues
Wei Chen, Tongguan Wang, Feiyue Xue, Junkai Li, Hui Liu · Sep 19, 2025 · Citations: 0
- Quantifying Genuine Awareness in Hallucination Prediction Beyond Question-Side Shortcuts
Yeongbin Seo, Dongha Lee, Jinyoung Yeo · Sep 18, 2025 · Citations: 0
However, we argue that the reported performance to date reflects not only a model's genuine awareness of its internal information, but also awareness derived purely from question-side information (e.g., benchmark hacking).
- Evolving Language Models without Labels: Majority Drives Selection, Novelty Promotes Variation
Yujun Zhou, Zhenwen Liang, Haolin Liu, Wenhao Yu, Kishan Panaganti · Sep 18, 2025 · Citations: 0
Large language models (LLMs) are increasingly trained with reinforcement learning from verifiable rewards (RLVR), yet real-world deployment demands models that can self-improve without labels or external judges.
- ATTS: Asynchronous Test-Time Scaling via Conformal Prediction
Jing Xiong, Qiujiang Chen, Fanghua Ye, Zhongwei Wan, Chuanyang Zheng · Sep 18, 2025 · Citations: 0
Abstract shows limited direct human-feedback or evaluation-protocol detail; use as adjacent methodological context.
- Llama-Mimi: Exploring the Limits of Flattened Speech Language Modeling
Issa Sugiura, Shuhei Kurita, Yusuke Oda, Ryuichiro Higashinaka · Sep 18, 2025 · Citations: 0
Abstract shows limited direct human-feedback or evaluation-protocol detail; use as adjacent methodological context.
- GeoResponder: Towards Building Geospatial LLMs for Time-Critical Disaster Response
Ahmed El Fekih Zguir, Ferda Ofli, Muhammad Imran · Sep 18, 2025 · Citations: 0
Extensive evaluations across four topologically distinct cities and diverse tasks demonstrate that GeoResponder significantly outperforms both state-of-the-art foundation models and domain-specific baselines.
- Frame Sampling Strategies Matter: A Benchmark for small vision language models
Marija Brkic, Anas Filali Razzouki, Yannis Tevissen, Khalil Guetari, Mounim A. El Yacoubi · Sep 18, 2025 · Citations: 0
- ClearFairy: Capturing Creative Workflows through Decision Structuring, In-Situ Questioning, and Rationale Inference
Kihoon Son, DaEun Choi, Tae Soo Kim, Young-Ho Kim, Sangdoo Yun · Sep 18, 2025 · Citations: 0
Critique Edit
Furthermore, exploratory applications demonstrate that captured steps can enhance generative AI agents in Figma, yielding predictions better aligned with professionals and producing coherent outcomes.
- Bridging Past and Future: Distribution-Aware Alignment for Time Series Forecasting
Yifan Hu, Jie Yang, Tian Zhou, Peiyuan Liu, Yujin Tang · Sep 17, 2025 · Citations: 0
- Reasoning Efficiently Through Adaptive Chain-of-Thought Compression: A Self-Optimizing Framework
Kerui Huang, Shuhan Liu, Xing Hu, Tongtong Xu, Lingfeng Bao · Sep 17, 2025 · Citations: 0
To investigate these trade-offs, we conduct an empirical study based on code generation benchmarks.
- Masked Diffusion Models as Energy Minimization
Sitong Chen, Shen Nie, Jiacheng Sun, Zijin Feng, Zhenguo Li · Sep 17, 2025 · Citations: 0
Experiments on synthetic and real-world benchmarks demonstrate that our energy-inspired schedules outperform hand-crafted baselines, particularly in low-step sampling settings.
- A Simple and Efficient Jailbreak Method Exploiting LLMs' Helpfulness
Xuan Luo, Yue Wang, Zefeng He, Geng Tu, Jing Li · Sep 17, 2025 · Citations: 0
Red Team
This study reveals a critical safety blind spot in modern LLMs: learning-style queries, which closely resemble ordinary educational questions, can reliably elicit harmful responses.
- See, Think, Act: Teaching Multimodal Agents to Effectively Interact with GUI by Identifying Toggles
Zongru Wu, Rui Mao, Zhiyuan Tian, Pengzhou Cheng, Tianjie Ju · Sep 17, 2025 · Citations: 0
To address the challenge, we propose State-aware Reasoning (StaR), a multimodal reasoning method that enables agents to perceive the current toggle state, infer the desired state from the instruction, and act accordingly.
- Linear probes rely on textual evidence: Results from leakage mitigation studies in language models
Gerard Boxo, Aman Neelappa, Shivam Raval · Sep 16, 2025 · Citations: 0
- ReSum: Unlocking Long-Horizon Search Intelligence via Context Summarization
Xixi Wu, Kuan Li, Yida Zhao, Liwen Zhang, Litu Ou · Sep 16, 2025 · Citations: 0
Tool Use
Large Language Model (LLM)-based web agents excel at knowledge-intensive tasks but face a fundamental conflict between the need for extensive exploration and the constraints of limited context windows.
- From Next Token Prediction to (STRIPS) World Models
Carlos Núñez-Molina, Vicenç Gómez, Hector Geffner · Sep 16, 2025 · Citations: 0
- DoubleAgents: Human-Agent Alignment in a Socially Embedded Workflow
Tao Long, Xuanming Zhang, Sitong Wang, Zhou Yu, Lydia B Chilton · Sep 16, 2025 · Citations: 0
Pairwise Preference
We present DoubleAgents, a system for human-agent alignment in coordination tasks, grounded in distributed cognition.
- Neural-Quantum-States Impurity Solver for Quantum Embedding Problems
Yinzhanghao Zhou, Tsung-Han Lee, Ao Chen, Nicola Lanatà, Hong Guo · Sep 15, 2025 · Citations: 0
- The AI Memory Gap: Users Misremember What They Created With AI or Without
Tim Zindulka, Sven Goller, Daniela Fernandes, Robin Welsch, Daniel Buschek · Sep 15, 2025 · Citations: 0
Our findings reveal a significant gap in memory: After AI use, the odds of correct attribution dropped, with the steepest decline in mixed human-AI workflows, where either the idea or elaboration was created with AI.
- Collaborative Document Editing with Multiple Users and AI Agents
Florian Lehmann, Krystsina Shauchenka, Daniel Buschek · Sep 15, 2025 · Citations: 0
We propose integrating AI agents directly into collaborative writing environments.
- PeruMedQA: Benchmarking Large Language Models (LLMs) on Peruvian Medical Exams -- Dataset Construction and Evaluation
Rodrigo M. Carrillo-Larco, Jesus Lovón Melgarejo, Manuel Castillo-Cara, Gusseppe Bravo-Rocca · Sep 15, 2025 · Citations: 0
Abstract shows limited direct human-feedback or evaluation-protocol detail; use as adjacent methodological context.
- RAPTOR: A Foundation Policy for Quadrotor Control
Jonas Eschmann, Dario Albani, Giuseppe Loianno · Sep 15, 2025 · Citations: 0
Demonstrations Long Horizon
Humans are remarkably data-efficient when adapting to new unseen conditions, like driving a new car.
- Learning to Optimize Multi-Objective Alignment Through Dynamic Reward Weighting
Yining Lu, Zilong Wang, Shiyang Li, Xin Liu, Changlong Yu · Sep 14, 2025 · Citations: 0
Pairwise Preference
This limitation becomes especially critical in online preference alignment for large language models.
- CogniAlign: Survivability-Grounded Multi-Agent Moral Reasoning for Safe and Transparent AI
Hasin Jawad Ali, Ilhamul Azam, Ajwad Abrar, Md. Kamrul Hasan, Hasan Mahmud · Sep 14, 2025 · Citations: 0
Multi Agent
The challenge of aligning artificial intelligence (AI) with human values persists due to the abstract and often conflicting nature of moral principles and the opacity of existing approaches.
- No Answer Needed: Predicting LLM Answer Accuracy from Question-Only Linear Probes
Iván Vicente Moreno Cencerrado, Arnau Padrés Masdemont, Anton Gonzalvez Hawthorne, David Demitri Africa, Lorenzo Pacchiardi · Sep 12, 2025 · Citations: 0
Abstract shows limited direct human-feedback or evaluation-protocol detail; use as adjacent methodological context.
- Incongruent Positivity: When Miscalibrated Positivity Undermines Online Supportive Conversations
Leen Almajed, Abeer ALdayel · Sep 12, 2025 · Citations: 0
We examine this phenomenon of incongruent positivity as miscalibrated expressions of positive support in both human and LLM generated responses.
- Human Psychometric Questionnaires Mischaracterize LLM Psychology: Evidence from Generation Behavior
Woojung Song, Dongmin Choi, Yoonah Park, Jongwook Han, Eun-Ju Lee · Sep 12, 2025 · Citations: 0
Rubric Rating
Psychological profiling of large language models (LLMs) using psychometric questionnaires designed for humans has become widespread.
- Linguistic trajectories of bipolar disorder on social media
Laurin Plank, Armin Zlomuzica · Sep 12, 2025 · Citations: 0
Abstract shows limited direct human-feedback or evaluation-protocol detail; use as adjacent methodological context.
- LoFT: Parameter-Efficient Fine-Tuning for Long-tailed Semi-Supervised Learning in Open-World Scenarios
Zhiyuan Huang, Jiahao Chen, Bing Su · Sep 12, 2025 · Citations: 0
Experimental results on multiple benchmarks demonstrate that our method achieves superior performance.
- The Illusion of Diminishing Returns: Measuring Long Horizon Execution in LLMs
Akshit Sinha, Arvindh Arun, Shashwat Goel, Steffen Staab, Jonas Geiping · Sep 11, 2025 · Citations: 0
- OTESGN: Optimal Transport-Enhanced Syntactic-Semantic Graph Networks for Aspect-Based Sentiment Analysis
Xinfeng Liao, Xuanqi Chen, Lianxi Wang, Jiahuan Yang, Zhuowei Chen · Sep 10, 2025 · Citations: 0
- Evolution and compression in LLMs: On the emergence of human-aligned categorization
Nathaniel Imel, Noga Zaslavsky · Sep 9, 2025 · Citations: 0
Converging evidence suggests that human systems of semantic categories achieve near-optimal compression via the Information Bottleneck (IB) complexity-accuracy tradeoff.
- SimpleQA Verified: A Reliable Factuality Benchmark to Measure Parametric Knowledge
Lukas Haas, Gal Yona, Giovanni D'Antonio, Sasha Goldshtein, Dipanjan Das · Sep 9, 2025 · Citations: 0
We introduce SimpleQA Verified, a 1,000-prompt benchmark for evaluating Large Language Model (LLM) short-form factuality based on OpenAI's SimpleQA.
- FHIR-RAG-MEDS: Integrating HL7 FHIR with Retrieval-Augmented Large Language Models for Enhanced Medical Decision Support
Yildiray Kabak, Gokce B. Laleci Erturkmen, Mert Gencturk, Tuncay Namli, A. Anil Sinaci · Sep 9, 2025 · Citations: 0
Abstract shows limited direct human-feedback or evaluation-protocol detail; use as adjacent methodological context.
- MedicalPatchNet: A Patch-Based Self-Explainable AI Architecture for Chest X-ray Classification
Patrick Wienholt, Christiane Kuhl, Jakob Nikolas Kather, Sven Nebelung, Daniel Truhn · Sep 9, 2025 · Citations: 0
Abstract shows limited direct human-feedback or evaluation-protocol detail; use as adjacent methodological context.
- LLM Analysis of 150+ years of German Parliamentary Debates on Migration Reveals Shift from Post-War Solidarity to Anti-Solidarity in the Last Decade
Aida Kostikova, Ole Pütz, Steffen Eger, Olga Sabelfeld, Benjamin Paassen · Sep 8, 2025 · Citations: 0
We first provide a comprehensive evaluation of multiple LLMs, analyzing the effects of model size, prompting strategies, fine-tuning, historical versus contemporary data, and systematic error patterns.
- Index-Preserving Lightweight Token Pruning for Efficient Document Understanding in Vision-Language Models
Jaemin Son, Sujin Choi, Inyong Yun · Sep 8, 2025 · Citations: 0