Search: “reasoning”

Safe Error Correction for Language Models: Frozen-Base Adjustment with Capability Preservation

arXiv:2609.16145v1 Announce Type: new Abstract: We study a practical question: can a small correction module fix errors in a frozen language model's outputs without degrading its base capabilities? We

Content type

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

Research

Shared-Prefix KV Reuse Across Standard LoRA Adapters: Quality and Serving Tradeoffs

DGX agent

arXiv:2609.17109v1 Announce Type: new Abstract: A common small-model deployment runs one shared backbone with several LoRA specialists that answer over the same context. Serving them naively re-prefil

researcharxiv-cs-ai

16 Sept 2026

SlotDiT: Object-Centric Representations for Diffusion Transformers

arXiv:2609.17414v1 Announce Type: new Abstract: Text-conditioned latent diffusion models perform strongly in video generation and are promising backbones for robotic applications. However, existing ap

safetyarxiv-cs-cv

Style-Debiased DPO: Updating LLM Knowledge with Factuality-Aware Synthetic Preference Data

model-releasesarxiv-cs-cl

arXiv:2609.16532v1 Announce Type: new Abstract: Continued pretraining (CPT) with data augmentation such as paraphrasing can store inside a large language model (LLM) the knowledge of a small source co

Symbolic Separation: Grounding Deep Agents in Knowledge Graphs for Trustworthy Operational Data Analytics

arXiv:2609.17107v1 Announce Type: new Abstract: Generative AI promises natural language access to the massive numerical telemetry of data centers and Industry 4.0 installations, yet text-to-query and

TIAO: Token Importance-Aware Policy Optimization for Text Summarization

model-releasesarxiv-cs-cl

arXiv:2609.16748v1 Announce Type: new Abstract: Text summarization requires models to condense content while preserving key qualities such as consistency and coherence. Large language models (LLMs) ha

Towards Scalable RLVR: Multimodal Instruction Following Data Synthesis and Distillation

arXiv:2609.16059v1 Announce Type: new Abstract: Multimodal instruction following (MMIF) is crucial for building generalist agents. However, current training paradigms rely heavily on Supervised Fine-T

safetyarxiv-cs-cl

UDAV: Uncertainty-Driven Adaptive VLM Waypoint Planner

arXiv:2609.16368v1 Announce Type: cross Abstract: Vision-language models (VLMs) can generate routes directly from aerial imagery for off-road navigation, but their predictions provide no indication of

Local Ai

When Confidence Signals Disagree: Local and Global Confidence in Autoregressive Language Models

arXiv:2609.16933v1 Announce Type: cross Abstract: Modern predictive systems expose multiple quantities that are commonly interpreted as measures of confidence. However, these quantities can summarize

local-aiarxiv-cs-ai

Where Should the KV Cache Live? Placement Policies Across GPU, CPU, and SSD for Long-Lived Sessions

arXiv:2609.16215v1 Announce Type: new Abstract: GPU high bandwidth memory is scarce and expensive, and KV caches consume much of it as chats, agent loops, and document question answering accumulate st

safetyarxiv-cs-ai

World Models for Embodied Intelligence: From Plausible to Controllable to Actionable

arXiv:2609.16697v1 Announce Type: cross Abstract: World models connect perception and decision-making in embodied intelligence by maintaining hidden state, anticipating consequences, comparing interve

3D Gait-Based Autism Classification Using Attention-Enhanced Deep Learning with Cross-Fold Statistical Stability Analysis

arXiv:2609.14159v1 Announce Type: new Abstract: Autism Spectrum Disorder (ASD) is a neurodevelopmental condition whose early diagnosis remains challenging because conventional clinical assessments are

Tutorials

A Chosen Future Can Still Be Rewritten: Causal Writability in Video Models

arXiv:2609.15980v1 Announce Type: cross Abstract: When a video model generates physically incorrect motion, did it fail to learn the correct motion, or did it learn it but fail to use it? We show the

tutorialsarxiv-cs-cv

A Comprehensive Review of Multimodal Facial State Analysis: Tasks, Methods, and Resources

arXiv:2609.13255v1 Announce Type: new Abstract: Facial state analysis plays a crucial role in understanding human expressions, psychological modeling, and human computer interaction. Traditional unimo

researcharxiv-cs-cv

A Hybrid Agentic AI Framework for Intelligent Supply Chain Analytics

arXiv:2609.13561v1 Announce Type: new Abstract: Efficient utilization of supply chain analytics for decision making remains a significant challenge for planners, as critical tasks such as database que

Applications

AlgoRAG: Retrieval-Augmented Generation for Theoretical Computer Science Education -- A Comprehensive Evaluation Framework for Algorithm Analysis and Complexity Theory

arXiv:2609.14572v1 Announce Type: cross Abstract: Teaching abstract theoretical computer science (TCS) concepts such as algorithm analysis and complexity theory is challenging because students must ha

applicationsarxiv-cs-ai

ATTRICITE: Training an Open 4B Model for Citation Recovery toward Faithful Attribution

arXiv:2609.14248v1 Announce Type: cross Abstract: Faithful citation attribution begins with identifying the intended source for a scientific claim. We study this source-identification capability throu

Bridging the Modality Gap in Long-Form Clinical Audio: A Comparative Study of Lightweight and Heavyweight End-to-End SOAP Generation

arXiv:2609.14467v1 Announce Type: cross Abstract: Automating clinical documentation from long-form doctor-patient conversations remains challenging for modern audio-language models. While cascaded ASR

Calibrating Interpretability Instruments Before Trusting Their Verdicts

arXiv:2609.14754v1 Announce Type: cross Abstract: Causal claims about large language model (LLM) internals rest on measurements. Those might include a projection, a cosine, an ablation delta, or an in

Can We Trust the Judges? Validation of Factuality Evaluation Methods via Answer Perturbation

arXiv:2609.15561v1 Announce Type: new Abstract: Evaluating the factual correctness of large language models (LLMs) is vital for many applications. But are our evaluation tools themselves trustworthy?

researcharxiv-cs-cl

Carbon-Aware Routing for Function Calling in Edge-Cloud LLM Systems

arXiv:2609.13559v1 Announce Type: new Abstract: Large Language Models (LLMs) with function-calling capabilities are becoming critical for modern agentic AI systems. Nevertheless, current deployments t

ChatGPT Images 2.5 in the Wild: A Launch-Period Dataset and Detector Evaluation

arXiv:2609.15100v1 Announce Type: cross Abstract: An image tool can change its underlying generator while retaining its public name, making version attribution from online posts ambiguous. We study th

Cost Characterization of Vertically Partitioned Federated Knowledge Graphs

arXiv:2609.13664v1 Announce Type: new Abstract: Knowledge graphs are increasingly distributed across autonomous organizations that share an entity space but own disjoint subsets of relations, forming

DecompressionLM: Deterministic, Diagnostic, and Zero-Shot Concept Graph Extraction from Language Models

arXiv:2602.00377v3 Announce Type: replace Abstract: Existing knowledge probing methods rely on pre-defined queries, limiting extraction to known concepts. We introduce DecompressionLM, a stateless fra

researcharxiv-cs-cl

Decoupling Error Attribution in Cloud-Native Graph-RAG: A Data Integrity Diagnostic Framework

arXiv:2609.13324v1 Announce Type: cross Abstract: Graph-RAG systems often assume pristine data quality, overlooking the severe impact of perturbations in cloud-native databases. This paper proposes a

Designing UNICORN: a Unified Benchmark for Imaging in Computational Pathology, Radiology, and Natural Language

arXiv:2603.02790v2 Announce Type: replace Abstract: Foundation models are changing the way we develop medical artificial intelligence. By learning broadly generalizable features across diverse data mo

Diagnosing Faults in Reinforcement Learning Simulators and World Models with Canonical Polynomial Invariants

arXiv:2609.13194v1 Announce Type: cross Abstract: A large literature builds physical structure into learned dynamics on the premise that models respecting the underlying physics predict better. We tes

Do Not Restart: Residual Completion for Stateful Agent Handoffs

arXiv:2609.13800v1 Announce Type: new Abstract: Routing and cascades reduce tool-agent cost by transferring control across models, but stateful handoffs must preserve accepted choices, realized effect

Dynamic Semantic Compression for Efficient Latent-Space Inference in Large Language Models

arXiv:2609.15338v1 Announce Type: cross Abstract: Large Language Models (LLMs) primarily perform inference at the token level, resulting in substantial memory overhead and compromised computational ef

ECAS: An Edge-Controlled Agentic System for Validation-Gated Scientific Application Execution

arXiv:2609.14211v1 Announce Type: cross Abstract: Scientific applications increasingly rely on high-performance computing (HPC), yet translating a scientist's high-level goal into a correct target-sca

safetyarxiv-cs-ai

Efficiency Hallucination: Formalizing and Measuring Behavioral Calibration in LLM-Based Code Optimization

arXiv:2609.14839v1 Announce Type: cross Abstract: The integration of Large Language Models (LLMs) into automated code optimization introduces a critical reliability risk we term the Efficiency Halluci

Enemray: Toward Capable Language Models for Hassaniya

arXiv:2609.14829v1 Announce Type: cross Abstract: We introduce Enemray, a Hassaniya-centric language model that enables general-purpose interaction in Hassaniya. Enemray is trained around a stability-

Evaluating LLM-Generated Rules for Heart Disease Prediction

model-releasesarxiv-cs-lg

arXiv:2609.13192v1 Announce Type: new Abstract: This study compares traditional machine learning models and Large Language Model (LLM)-generated rule-based systems for heart disease prediction using t

Externalizing Requirement-to-Repair Artifacts as Observable Traces for LLM-Based Program Repair

arXiv:2609.14913v1 Announce Type: new Abstract: Repository-level repair requires not only correct patches but also inspectable records that explain how issue requirements are translated into code chan

From Vision to Harvest: Benchmarking Vision-Language Models for Multi-Arm Robotic Fruit Harvesting

arXiv:2609.13606v1 Announce Type: cross Abstract: Multi-arm robotic harvesting offers a promising path to improve harvesting efficiency and reduce reliance on manual labor. However, practical deployme

GEAR: From Dynamic Encoding to Dynamic Activation in Social Trajectory Prediction

arXiv:2609.13778v1 Announce Type: cross Abstract: Human trajectory prediction requires modeling both individual motion patterns and social interactions among agents. Existing methods have made substan

safetyarxiv-cs-ai

Tutorials

Generalized Correctness Models: Learning Calibrated and Model-Agnostic Correctness Predictors from Historical Patterns

arXiv:2509.24988v2 Announce Type: replace-cross Abstract: Generating accurate and calibrated confidence estimates is critical for deploying LLMs in high-stakes or user-facing applications, and remains

tutorialsarxiv-cs-ai

Geometric Signatures of Conceptual Reorganization: A Counterfactual Embedding Framework for Detecting Scientific Revolutions

arXiv:2609.14917v1 Announce Type: cross Abstract: We introduce document embedding geometry as a quantitative observable of conceptual reorganization and develop a counterfactual ablation framework for

researcharxiv-cs-cl

KnowBench: Effort Reduction as a Unified, Deployment-Grounded Benchmark for Clinical AI

arXiv:2609.15794v1 Announce Type: new Abstract: Clinical AI systems are evaluated with instruments built for research settings (reference-based similarity metrics and expert rubric panels) that measur

Learning from Reliable Negatives: Confidence-Anchored Test-Time Adaptation for GUI Grounding

arXiv:2609.15307v1 Announce Type: new Abstract: Graphical User Interface (GUI) grounding is essential for autonomous agents to map natural language instructions to precise screen coordinates. However,

agentsarxiv-cs-cv

LG-VLN: A Zero-Shot Vision-and-Language Navigation Framework with LangGraph State Orchestration

arXiv:2609.15098v1 Announce Type: new Abstract: Continuous-environment vision-and-language navigation (VLN-CE) requires interpreting natural-language instructions in unseen 3D environments and executi

LLM Probability Concentration: How Alignment Shrinks the Generative Horizon

arXiv:2506.17871v4 Announce Type: replace-cross Abstract: Despite their impressive capabilities, aligned large language models (LLMs) often generate outputs that lack diversity. What drives this consi

LongAgent: History-Guided Agentic Search for Longitudinal Outcome Prediction

arXiv:2609.15859v1 Announce Type: new Abstract: Extracting informative representations from longitudinal data that can predict future outcomes remains a critical challenge in medicine. Medical dataset

Look Before You Leap: Factual Decoding with Internal Attribution Signals

arXiv:2609.15745v1 Announce Type: cross Abstract: Hallucination remains a critical challenge in large language models (LLMs), where early factual errors compound through autoregressive generation in a

Lost-in-the-Middle in Long-Text Generation: Synthetic Dataset, Evaluation Framework, and Mitigation

arXiv:2503.06868v2 Announce Type: replace-cross Abstract: Existing long-text generation methods produce lengthy outputs from short inputs, leaving long-input-to-long-output generation underexplored. A

Machine Unlearning for Speech Question Answering in Large Audio-Language Models

arXiv:2609.13195v1 Announce Type: cross Abstract: Large Audio-Language Models (LALMs) have recently shown strong capabilities in speech understanding and question answering (QA), but they also inherit

safetyarxiv-cs-cl

Map Users and Mapmakers: The Scope of Cognitive Attribution from Acquired Representations

arXiv:2609.13879v1 Announce Type: new Abstract: An acquired representation can enlarge a system's cognitive repertoire without transferring the capacities exercised in producing that representation. T

Applications

Medical Knowledge Simplification for Patients in the Era of LLMs: A Case Study on Diabetes

arXiv:2609.15129v1 Announce Type: new Abstract: Complex medical information is often difficult for patients to understand, making effective medical knowledge simplification essential for improving pat

applicationsarxiv-cs-ai