Instant research discovery
Search and browse arXiv CS/AI/ML papers, enriched with AI-generated insights.
Generate novel research ideas grounded in real arXiv papers with Brainstorm.
PreviousNext
| Paper | Published | Area | Tags | Intel | Citations |
|---|---|---|---|---|---|
Search and browse arXiv CS/AI/ML papers, enriched with AI-generated insights.
Generate novel research ideas grounded in real arXiv papers with Brainstorm.
| Paper | Published | Area | Tags | Intel | Citations |
|---|---|---|---|---|---|
Search and browse arXiv CS/AI/ML papers, enriched with AI-generated insights.
Generate novel research ideas grounded in real arXiv papers with Brainstorm.
| Paper | Published | Area | Tags | Intel | Citations |
|---|---|---|---|---|---|
| Paper | Published | Area | Tags | Intel | Citations |
|---|---|---|---|---|---|
| What Does Post-Training Change in Multilingual Reasoning? Caesar Wu, Gr茅goire Danoy, Hongyang Li, Pascal Bouvry Published: 2026-09-29Area: cs.CLCitations: - Tags: ai-safety, cscl, preprint | 2026-09-29 | cs.CL | ai-safety, cscl, preprint | - | - |
| Which Models Work Well Together? Measuring Heterogeneity for LLM Team Selection Hengsong Liu, Jing Liu, Jingyu Zhang, Juncen Guo Published: 2026-09-29Area: cs.CLCitations: - Tags: ai-safety, cscl, preprint | 2026-09-29 | cs.CL | ai-safety, cscl, preprint | E10 / R7 (89%) | - |
| AgentHop: A Diagnostic Benchmark for Agentic Multi-Hop Scientific Question Answering Chanhee Park, Heuiseok Lim, Hyeonseok Moon, Jeongho Yoon Published: 2026-09-28Area: cs.CLCitations: - Tags: ai-safety, cscl, preprint | 2026-09-28 | cs.CL | ai-safety, cscl, preprint | - | - |
| Almieyar: A Culturally Grounded Benchmark for Multi-Dialect Arabic Speech Recognition Ahd Al Jumah, Ahmed Ezzat, Aisha Hamad M A Al-Naimi, Alhanouf Abdulraqib Published: 2026-09-28Area: cs.CLCitations: - Tags: ai-safety, cscl, preprint | 2026-09-28 | cs.CL | ai-safety, cscl, preprint | - | - |
| Almost Human, Except When It Matters: VoxParity and the Decisions a Voice Should Change Bhavik Mangla Published: 2026-09-28Area: cs.CLCitations: 50 Tags: ai-safety, cscl, preprint | 2026-09-28 | cs.CL | ai-safety, cscl, preprint | E10 / R6 (89%) | 50 |
| A mechanistic study of language model introspection Jiahong Zou, Lingkai Kong, Tonghan Wang, Xiangkun Sun Published: 2026-09-28Area: cs.CLCitations: - Tags: ai-safety, cscl, preprint | 2026-09-28 | cs.CL | ai-safety, cscl, preprint | - | - |
| AwarenessBench: Assessing Cognitive Capabilities of Language Models Briana Zhang, Haoyuan Shi, Kyle Xue Chen, Peiran Yang Published: 2026-09-28Area: cs.CLCitations: - Tags: ai-safety, cscl, preprint | 2026-09-28 | cs.CL | ai-safety, cscl, preprint | - | - |
| Beyond Token Alignment: Event Completion for Cross-Tokenizer On-Policy Distillation Jiacheng Liu, Jingwei Song, Linfeng Zhang, Qituan Zhang Published: 2026-09-28Area: cs.CLCitations: - Tags: ai-safety, alignment-training, cscl, preprint | 2026-09-28 | cs.CL | ai-safety, alignment-training, cscl, preprint | - | - |
| Beyond Token Scale: Chunk-Level Sparse Autoencoders for Reliable Semantic Feature Discovery Difan Zou, TingHao YU, Xu Wang, Yifan Yang Published: 2026-09-28Area: cs.CLCitations: - Tags: ai-safety, cscl, preprint | 2026-09-28 | cs.CL | ai-safety, cscl, preprint | - | - |
| Calibrated to Whom? Persona and Language Effects on Cultural Values in JEV Abdulaziz Asseri, Bushra Asseri Published: 2026-09-28Area: cs.CLCitations: - Tags: ai-safety, cscl, preprint | 2026-09-28 | cs.CL | ai-safety, cscl, preprint | - | - |
| Causal and Interpretable Structures in LLM Compositional Tasks Christopher J. Earls, Gurbir Arora, Jiajun Bao, Rapha毛l Sarfati Published: 2026-09-28Area: cs.CLCitations: - Tags: ai-safety, cscl, preprint | 2026-09-28 | cs.CL | ai-safety, cscl, preprint | E11 / R7 (95%) | - |
| CineSubBench: Evaluating LLMs on Long-Form Narrative and Cultural Understanding from Multilingual Movie Subtitles Davood Rafiei, Mir Tafseer Nayeem, Susmoy Chakraborty Published: 2026-09-28Area: cs.CLCitations: - Tags: ai-safety, cscl, preprint | 2026-09-28 | cs.CL | ai-safety, cscl, preprint | - | - |
| Coherence-Aware Distributional Evaluation of Open-Ended Text Generation Haoming Liu, Hongyi Wen, Jinnuo Liu, Junhao Zhu Published: 2026-09-28Area: cs.CLCitations: - Tags: ai-safety, cscl, preprint, safety-evaluation | 2026-09-28 | cs.CL | ai-safety, cscl, preprint, safety-evaluation | - | - |
| Coherence-Aware Distributional Evaluation of Open-Ended Text Generation Haoming Liu, Hongyi Wen, Jinnuo Liu, Junhao Zhu Published: 2026-09-28Area: cs.CLCitations: - Tags: ai-safety, cscl, preprint, safety-evaluation | 2026-09-28 | cs.CL | ai-safety, cscl, preprint, safety-evaluation | E8 / R6 (91%) | - |
| Dr.Credit: Rubric-Grounded Process Credit Assignment for Deep Research Agents Jiaxin Guo, Kun Ding, Pengjun Xie, Shen Huang Published: 2026-09-28Area: cs.CLCitations: - Tags: ai-safety, cscl, preprint | 2026-09-28 | cs.CL | ai-safety, cscl, preprint | - | - |
| Echoes of Deeds: Moral History Can Shape and Steer LLM Behavioral Choices Andrea Tagarelli, Lucio La Cava Published: 2026-09-28Area: cs.CLCitations: - Tags: ai-safety, cscl, preprint | 2026-09-28 | cs.CL | ai-safety, cscl, preprint | - | - |
| Evaluating Bounded Autonomy in Regulated Agentic AI: A Diagnostic Harness with Constitutional Rewards, Escalation Labels, and Runtime Governance Dipankar Sarkar Published: 2026-09-28Area: cs.CLCitations: - Tags: ai-safety, cscl, preprint | 2026-09-28 | cs.CL | ai-safety, cscl, preprint | E10 / R9 (92%) | - |
| FactorEngram: Factorized N-gram Memory with Basis-Level Gating for Language Models Bowen Yang, Hua Wu, Jingbo Zhou, Qinghong Miao Published: 2026-09-28Area: cs.CLCitations: - Tags: ai-safety, cscl, preprint | 2026-09-28 | cs.CL | ai-safety, cscl, preprint | - | - |
| Fair Fact-Checking: Closing the Cross-Lingual Gap in LLM Factual Judgement with RoSh Adrian Weller, Dongwon Lee, Fatemeh Seyedin, Mahmoudreza Babaei Published: 2026-09-28Area: cs.CLCitations: - Tags: ai-safety, cscl, preprint | 2026-09-28 | cs.CL | ai-safety, cscl, preprint | - | - |
| From Normative Frameworks to Alignment Data: Constructing and Evaluating SFT and Preference Data Ala AlFuqaha, Danish Naeem, Husrev Taha Sencar, Ji Lucas Published: 2026-09-28Area: cs.CLCitations: - Tags: ai-safety, alignment-training, cscl, preprint | 2026-09-28 | cs.CL | ai-safety, alignment-training, cscl, preprint | - | - |
| In-game Toxic Detection: Bi-directional Representations with Attention Residuals Yuanzhe Jia Published: 2026-09-28Area: cs.CLCitations: - Tags: ai-safety, cscl, preprint | 2026-09-28 | cs.CL | ai-safety, cscl, preprint | - | - |
| Just-In-Time Agent Memory with Runtime Agentic Research Bingyu Yan, Chaofan Li, Chaozhuo Li, Hongjin Qian Published: 2026-09-28Area: cs.CLCitations: - Tags: ai-safety, cscl, preprint | 2026-09-28 | cs.CL | ai-safety, cscl, preprint | - | - |
| Language Models Are "Insecure" Reporters Ahmed Imtiaz Humayun, Hongxiang Gu, Jenny Y. Huang, Jiameng Fan Published: 2026-09-28Area: cs.CLCitations: - Tags: ai-safety, cscl, preprint | 2026-09-28 | cs.CL | ai-safety, cscl, preprint | - | - |
| Less Sycophancy, Stronger Refusal? Lessons for AI Safety from Mechanistic Interpretability Difan Zou, Xuansheng Wu, Xu Wang Published: 2026-09-28Area: cs.CLCitations: - Tags: ai-safety, cscl, interpretability, preprint | 2026-09-28 | cs.CL | ai-safety, cscl, interpretability, preprint | - | - |
| LLMs are not stochastic parrots: Evidence for meaning-mediated abstraction from conlang-like tasks Calla G. Beauregard, Elisabeth Kollrack, Julia Witte Zimmerman, Kathryn Cramer Published: 2026-09-28Area: cs.CLCitations: 12 Tags: ai-safety, cscl, preprint | 2026-09-28 | cs.CL | ai-safety, cscl, preprint | E10 / R8 (96%) | 12 |
| LLMs are not stochastic parrots: Evidence for meaning-mediated abstraction from conlang-like tasks Calla G. Beauregard, Elisabeth Kollrack, Julia Witte Zimmerman, Kathryn Cramer Published: 2026-09-28Area: cs.CLCitations: - Tags: ai-safety, cscl, preprint | 2026-09-28 | cs.CL | ai-safety, cscl, preprint | - | - |
| LongPuzzleBench: Evaluating GUI Agents on Long-Horizon Visual Puzzles Ari Yu Zhang, Bingo Zhang, Chaozheng Wang, Genjian Li Published: 2026-09-28Area: cs.CLCitations: - Tags: ai-safety, cscl, preprint | 2026-09-28 | cs.CL | ai-safety, cscl, preprint | - | - |
| Mnemon: Raw Records, Fast Judgments, Slow Thoughts Guangren Wang Published: 2026-09-28Area: cs.CLCitations: 63 Tags: ai-safety, cscl, preprint | 2026-09-28 | cs.CL | ai-safety, cscl, preprint | E10 / R7 (93%) | 63 |
| MS-GLA: Multi-Scale Gated Linear Attention for Addressing Representational Bottlenecks via Multi-Temporal Resolution Anirudh Sankar, Prasoon Dev, Vasudeva Varma Published: 2026-09-28Area: cs.CLCitations: - Tags: ai-safety, cscl, preprint | 2026-09-28 | cs.CL | ai-safety, cscl, preprint | - | - |
| Multilinguality in Hybrid Attention LLMs Chenyuan Yang, Junlin Hu, Lucas Bandarkar, Mohsen Fayyaz Published: 2026-09-28Area: cs.CLCitations: - Tags: ai-safety, cscl, preprint | 2026-09-28 | cs.CL | ai-safety, cscl, preprint | - | - |