AI news archive.
10,661 articles filed under Research, newest first, from the outlets listed on the sources page.
- arXiv cs.AIresearchCineSubBench: Evaluating LLMs on Long-Form Narrative and Cultural Understanding from Multilingual Movie Subtitles
arXiv:2609.36218v1 Announce Type: cross Abstract: Large language models are increasingly evaluated in specialized domains such as law, medicine, software engineering, and cybersecurity, yet film remains comparatively und…
- arXiv cs.AIresearchMutually Adversarial Self-Training with Evolving Data for Unified Multimodal Models
arXiv:2609.36224v1 Announce Type: cross Abstract: Unified multimodal models (UMMs) combine image generation and visual understanding in a shared backbone. Since generation and understanding are inverse tasks, recent stud…
- arXiv cs.AIresearchThink Before You Restore: Risk-Aware Manchu Manuscript Restoration with Stroke-Guided Attention
arXiv:2609.36243v1 Announce Type: cross Abstract: Full-page blind restoration of historical Manchu manuscripts is challenging due to scarce annotations, unknown degradation regions, and fragile connected strokes. Generic…
- arXiv cs.AIresearchPopulation Fidelity: Evaluating Population Representativeness in LLMs
arXiv:2609.36253v1 Announce Type: cross Abstract: Large language models (LLMs) show considerable potential in simulating human attitudes and preferences. Prior work finds that LLM-generated responses can compress the ran…
- arXiv cs.AIresearchPaired Multimodal Scaling Laws
arXiv:2609.36263v1 Announce Type: cross Abstract: Existing multimodal scaling laws fit multimodality terms empirically after testing and never vary how much data is multimodally paired at fixed data budgets. We investiga…
- arXiv cs.AIresearchIn-Context Learning Amplifies a Latent Symbolic Circuit
arXiv:2609.36265v1 Announce Type: cross Abstract: Large language models can learn abstract rules from just a few in-context examples, but how their internal mechanisms activate as examples accumulate is not well understo…
- arXiv cs.AIresearchProofs Without Nominals: G\"odel's Ontological Argument, its Shallow Embedding, and the Open Questions of the Monatshefte Notes
arXiv:2609.36279v4 Announce Type: replace-cross Abstract: The shallow embedding of higher-order modal logic in classical higher-order logic, used in Benzm\"uller and Scott's Notes on G\"odel's and Scott's variants of the…
- arXiv cs.AIresearchHow Much Prompt Is Enough? A Blackbox Minimization of Few-Shots in LLMs
arXiv:2609.36289v1 Announce Type: cross Abstract: Prompts are the primary mechanism for directing the behavior of large language models (LLMs). Yet the internal structure and causal hierarchy of prompts remain poorly und…
- arXiv cs.AIresearchMoRE: Scaling mixture of experts with hardware-aware low-rank routing
arXiv:2609.36301v1 Announce Type: cross Abstract: Mixture-of-Experts (MoE) layers are central to frontier language models, and recent architectures push toward more and smaller experts. In this regime, the standard linea…
- arXiv cs.AIresearchPyroStack: A Multi-Band Spatio-Temporal Sub-Daily Dataset for Wildfires in the United States
arXiv:2609.36315v1 Announce Type: cross Abstract: Wildfires are an increasing hazard to ecosystems, air quality, and human systems, creating a growing need for datasets that support systematic development and evaluation…
- arXiv cs.AIresearchTraining LLMs to Verbalize Evaluation Awareness
arXiv:2609.36316v1 Announce Type: cross Abstract: Evaluation awareness (EA) can cause large language models (LLMs) to behave differently during audits than in deployment, yet measuring and accounting for EA remains chall…
- arXiv cs.AIresearchPeriodic Weak Spots: Phase Sensitivity from Chunked KV-Cache Compression
arXiv:2609.36322v1 Announce Type: cross Abstract: Chunked KV-cache compression reduces the memory and attention costs of long-context inference by compressing windows of consecutive tokens into fewer cache entries at a f…
- arXiv cs.AIresearchDecoyTrace: Toxic Decoys for Active Defense in Decentralized Federated Learning
arXiv:2609.36330v1 Announce Type: cross Abstract: Decentralized Federated Learning (DFL) eliminates the central aggregation server, reducing the single point of observation that traditional defenses against attacks rely…
- arXiv cs.AIresearchCalibrating One-Round Membership Inference with Neighbors
arXiv:2609.36331v1 Announce Type: cross Abstract: The state-of-the-art Membership Inference (MI) methods calibrate their signal separately for each example using reference models, auxiliary models trained to exclude the…
- arXiv cs.AIresearchATLAS: Aligned Transport of Latent Structure for Reliable World Model Planning
arXiv:2609.36333v2 Announce Type: replace-cross Abstract: Latent world models rely on representation geometry for planning, yet regularizing the latent marginal alone does not determine the state-to-state relationships u…
- arXiv cs.AIresearchRepresentation by Design in Generation: Cross-View Class-Token Alignment in Diffusion Transformers
arXiv:2609.36348v1 Announce Type: cross Abstract: Generative and representation learning remain asymmetrically connected: semantic representations are used to improve diffusion generation, whereas the models' own represe…
- arXiv cs.AIresearchExplainability from Training with Applications to TCR-Epitope Prediction
arXiv:2609.36354v1 Announce Type: cross Abstract: Deep learning models have achieved strong performance in artificial intelligence for science, yet their black-box nature limits our understanding of how they learn scient…
- arXiv cs.AIresearchCompress to Remember: Learning Compact Memory via On-Policy Distillation for Long Video Generation
arXiv:2609.36364v1 Announce Type: cross Abstract: Standard video generators do not natively compact historical context into reusable memory tokens. As generation continues, the growing history makes it increasingly diffi…
- arXiv cs.AIresearchCross-attention encoding models reveal dynamic spatiotemporal routing across human higher visual cortex
arXiv:2609.36366v1 Announce Type: cross Abstract: Understanding how the brain parses actions and events from time-varying natural inputs is a central challenge in neuroscience. Recent work has used deep neural network (D…
- arXiv cs.AIresearchAdaKerNet: Neural Kernel Decoding for Task-Adaptive Prediction with Multimodal Large Models
arXiv:2609.36368v1 Announce Type: cross Abstract: Large foundation models have been introduced with the promise of efficient adaptation to downstream tasks. Yet, under limited supervision, MLLMs, an important class of la…
- arXiv cs.AIresearchQuantization Enables Private Dense Retrieval against Malicious Service Providers
arXiv:2609.36376v1 Announce Type: cross Abstract: Dense retrieval, the key component of Retrieval Augmented Generation (RAG), retrieves the most relevant documents by comparing dense vector representations of queries and…
- arXiv cs.AIresearchWhere Should Physics Enter a Molecular Crystal Generator?
arXiv:2609.36398v1 Announce Type: cross Abstract: Generative models make molecular crystal structure prediction fast, but their samples still exhibit geometric and packing violations. Physics can be introduced during tra…
- arXiv cs.AIresearchCalibrated to Whom? Persona and Language Effects on Cultural Values in JEV
arXiv:2609.36399v2 Announce Type: replace-cross Abstract: Decision-only language models return a probability for every answer option instead of generating text, which makes them attractive as survey respondents and as ju…
- arXiv cs.AIresearchLonger Records, Broader Invariance: The Hidden Scaling Problem in Longitudinal Contrastive Learning
arXiv:2609.36409v1 Announce Type: cross Abstract: Longitudinal data are valuable because people change. Yet the objectives used to learn from these data can inadvertently erase that change. In person-level contrastive le…
- arXiv cs.AIresearchFineART: Fine-Grained Annotated Robotic Trajectory Dataset and Vision-Language-Action Model for Bimanual Manipulation
arXiv:2609.36416v3 Announce Type: replace-cross Abstract: Robots operating in real-world environments must often execute complex, multi-step bimanual tasks over long horizons rather than single, isolated actions. Current…
- arXiv cs.AIresearchOnline Versatile Incremental Learning: Towards Class and Domain-Agnostic Adaptation at Any Time
arXiv:2609.36442v1 Announce Type: cross Abstract: Continual learning enables vision systems to adapt to ever-changing data distributions. Despite significant advances, existing approaches fail to capture continuous and c…
- arXiv cs.AIresearchReliable Parallel Decoding in Masked Diffusion Language Models
arXiv:2609.36452v1 Announce Type: cross Abstract: Masked diffusion language models (MDLMs) can generate text efficiently by predicting multiple masked tokens in parallel, but predictions from the same forward pass are no…
- arXiv cs.AIresearchChannel-Dependent State Space Model for Multivariate Time Series Forecasting
arXiv:2609.36453v2 Announce Type: replace-cross Abstract: Multivariate time series forecasting (MTSF) is critical across many real-world domains. Existing deep learning approaches fall into two paradigms with distinct li…
- arXiv cs.AIresearchEmergent Tonal Structure in Learned Chord Embeddings and Its Relation to Tonal Tension
arXiv:2609.36460v1 Announce Type: cross Abstract: Several tonal pitch spaces and computational models have been proposed to analyze tonal structure in Western tonal music, many of them grounded in principles from music t…
- arXiv cs.AIresearchStaircase Policy: Streaming Inference for World-Action Models with Large Action Chunks
arXiv:2609.36471v1 Announce Type: cross Abstract: World-Action Models (WAMs) improve robotic manipulation by conditioning action generation on predicted future observations, but future prediction adds further inference o…
- arXiv cs.AIresearchSimilar Choices, Different Attention: Cross-Modal Associations in Humans and Vision-Language Models
arXiv:2609.36475v1 Announce Type: cross Abstract: Cross-modal associations are systematic pairings of features across modalities, such as the association of 'bouba' with round shapes and 'kiki' with sharp shapes. Prior w…
- arXiv cs.AIresearchGuard Models Are Overconfident Where Base Models Are Uncertain
arXiv:2609.36477v1 Announce Type: cross Abstract: Guard models are used as safety classifiers, with confidence scores driving downstream moderation decisions. We evaluate five guard models for prompt classification and f…
- arXiv cs.AIresearchQuantum Computing for Network Security Classification: Near-Term Classification and Long-Term Memory Efficiency
arXiv:2609.36479v1 Announce Type: cross Abstract: Quantum computing has already been explored in several network-security applications. However, how quantum computing may contribute to network-security classification in…
- arXiv cs.AIresearchBenchmarking Vision-Language Models on Synapse and Neuron Proofreading in Connectomics
arXiv:2609.36492v2 Announce Type: replace-cross Abstract: We benchmarked vision-language models (VLMs) on the decisions annotators take when inspecting electron microscopy images in connectomics: synapse proofreading (pr…
- arXiv cs.AIresearchInterBias-SV: Compound Conditions in Speaker Verification
arXiv:2609.36500v1 Announce Type: cross Abstract: Speaker verification systems encounter combinations of noise, channel distortion, and changes in speech. Evaluating each condition separately does not establish whether t…
- arXiv cs.AIresearchTowards Breaking the Learning System Wall Using Multimodal Tutoring Transcriptions
arXiv:2609.36502v1 Announce Type: cross Abstract: Past research using log data has faced the "learning system wall," whereby few methods exist for generalizing models of student learning across platforms. Increasingly, o…
- arXiv cs.AIresearchLarge-scale factor analysis shows machine intelligence is only partially interpretable
arXiv:2609.36515v1 Announce Type: cross Abstract: A common assumption in language model development is that cognitive abilities are organized around a general, domain-free intelligence factor, like fluid intelligence in…
- arXiv cs.AIresearchLIBERO-MAX: Do Robot Policies Adapt When the World Changes?
arXiv:2609.36518v1 Announce Type: cross Abstract: Robots must often continue a task after a target moves, the viewpoint shifts, or an obstacle appears, even though their earlier observations and committed actions reflect…
- arXiv cs.AIresearchReliability Testing of Medical Model Performance under Distributed Deployment
arXiv:2609.36525v1 Announce Type: cross Abstract: Distributed inference has become an indispensable part of deploying medical models under practical latency, memory, and throughput constraints. Although modern frameworks…
- arXiv cs.AIresearchDraftTrace: A Multi-View Analytics Environment for AI-Integrated Writing
arXiv:2609.36544v1 Announce Type: cross Abstract: Generative AI has changed how students produce writing assignments. The final artifact is no longer sufficient to understand the process through which it was produced. We…
- arXiv cs.AIresearchInteractive-Policy Distillation with Bidirectional Propose-and-Verify
arXiv:2609.36546v1 Announce Type: cross Abstract: On-policy distillation (OPD) trains a student model on its self-generated trajectories with dense token-level teacher feedback. However, naive OPD may suffer from teacher…
- arXiv cs.AIresearchSERA: Scale-Equalized Rollout Allocation for Maximum Likelihood Reinforcement Learning
arXiv:2609.36552v1 Announce Type: cross Abstract: Maximum Likelihood Reinforcement Learning (MaxRL) targets prompt-wise log-success and has shown strong performance on reasoning tasks. Under finite rollout budgets, howev…
- arXiv cs.AIresearchHow Medical VLMs Underutilize Their Vision Encoders: A Dermatology Perspective
arXiv:2609.36557v1 Announce Type: cross Abstract: Medical Vision-Language Models (VLMs) show significant promise for clinical image understanding, offering accurate diagnosis with interpretable reasoning. However, a crit…
- arXiv cs.AIresearchHiTS-CL: A Continual Learning Framework for Long-Horizon Temporal Knowledge Graph Extrapolation
arXiv:2609.36559v1 Announce Type: cross Abstract: Extrapolative temporal knowledge graph reasoning (TKGR) predicts future facts from historical snapshots. Most existing methods train once on an early prefix of the timeli…
- arXiv cs.AIresearchFM-ReID: Selective Competitive Token Routing for Object Re-Identification
arXiv:2609.36560v1 Announce Type: cross Abstract: Object re-identification (ReID) faces a recurring challenge: different identities can share highly similar global appearances, while the cues that distinguish them are lo…
- arXiv cs.AIresearchThinkingGuard: Decoding Implicit Hazards via Step-by-Step Risk Attribution in Multimodal Large Language Models
arXiv:2609.36562v1 Announce Type: cross Abstract: While Multimodal Large Language Models (MLLMs) are increasingly deployed in safety-critical domains, their reliability is threatened by multimodal implicit risks. Unlike…
- arXiv cs.AIresearchFrom Checkpoint Variation to Selection Gains in Supervised Fine-Tuning
arXiv:2609.36569v1 Announce Type: cross Abstract: Checkpoint selection is a routine decision in supervised fine-tuning (SFT): training produces multiple checkpoints, but only one is retained. Yet fixed-budget comparisons…
- arXiv cs.AIresearchLong-Term Memory-Guided Enhancement for Target Perception in Audio-Language Models
arXiv:2609.36577v1 Announce Type: cross Abstract: Audio large language models (ALLMs) can reason about the content of audio recordings to perform complex tasks. However, these capabilities usually collapse in real-world…
- arXiv cs.AIresearchFactorized Scheduling Principle: Learning Interpretable and Transferable Policies via Structured Additive Functions
arXiv:2609.36578v1 Announce Type: cross Abstract: Scheduling problems arise from repeatedly selecting one item from a set of candidates based on their states. These problems often reduce to assigning priority scores and…
- arXiv cs.AIresearchScaling Video Generation for Reasoning: At What Cost?
arXiv:2609.36599v1 Announce Type: cross Abstract: We study whether scaling video generation enables models to reason about hidden information from the past frames, and at what computational cost. Our controlled benchmark…