Search: “llm”

Modeling Human Perspectives with Socio-Demographic Representations

arXiv:2604.18069v1 Announce Type: new Abstract: Humans often hold different perspectives on the same issues. In many NLP tasks, annotation disagreement can reflect valid subjective perspectives. Model

Content type

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

MTSQL-R1: Towards Long-Horizon Multi-Turn Text-to-SQL via Agentic Training

DGX agent

arXiv:2510.12831v3 Announce Type: replace Abstract: Multi-turn Text-to-SQL aims to translate a user's conversational utterances into executable SQL while preserving dialogue coherence and grounding to

model-releasesarxiv-cs-cl

21 Apr 2026

Local Ai

Network-wide Freeway Traffic Estimation Using Sparse Sensor Data: A Dirichlet Graph Auto-Encoder Approach

arXiv:2503.15845v2 Announce Type: replace Abstract: Network-wide Traffic State Estimation (TSE), which aims to infer a complete image of network traffic states with sparsely deployed sensors, plays a

local-aiarxiv-cs-lg

Oblivion: Self-Adaptive Agentic Memory Control through Decay-Driven Activation

arXiv:2604.00131v2 Announce Type: replace Abstract: Human memory adapts through selective forgetting: experiences become less accessible over time but can be reactivated by reinforcement or contextual

agentsarxiv-cs-cl

Online Conformal Prediction with Adversarial Semi-bandit Feedback via Regret Minimization

arXiv:2604.17984v1 Announce Type: new Abstract: Uncertainty quantification is crucial in safety-critical systems, where decisions must be made under uncertainty. In particular, we consider the problem

safetyarxiv-cs-lg

PARM: Pipeline-Adapted Reward Model

arXiv:2604.18327v1 Announce Type: cross Abstract: Reward models (RMs) are central to aligning large language models (LLMs) with human preferences, powering RLHF and advanced decoding strategies. While

PaTaRM: Bridging Pairwise and Pointwise Signals via Preference-Aware Task-Adaptive Reward Modeling

arXiv:2510.24235v3 Announce Type: replace Abstract: Reward models (RMs) are central to reinforcement learning from human feedback (RLHF), providing the critical supervision signals that align large la

safetyarxiv-cs-lg

Physics-Informed Neural Networks for Biological 2D{+}t Reaction-Diffusion Systems

arXiv:2604.18548v1 Announce Type: new Abstract: Physics-informed neural networks (PINNs) provide a powerful framework for learning governing equations of dynamical systems from data. Biologically-info

applicationsarxiv-cs-lg

Preparation of Fractal-Inspired Computational Architectures for Automated Neural Design Exploration

arXiv:2511.07329v3 Announce Type: replace-cross Abstract: It introduces FractalNet, a fractal-inspired computational architectures for advanced large language model analysis that mainly challenges mod

researcharxiv-cs-cv

Privacy Collapse: Benign Fine-Tuning Can Break Contextual Privacy in Language Models

arXiv:2601.15220v2 Announce Type: replace Abstract: We identify a novel phenomenon in language models: benign fine-tuning of frontier models can lead to privacy collapse. We find that diverse, subtle

Process Reward Models Meet Planning: Generating Precise and Scalable Datasets for Step-Level Rewards

arXiv:2604.17957v1 Announce Type: new Abstract: Process Reward Models (PRMs) have emerged as a powerful tool for providing step-level feedback when evaluating the reasoning of Large Language Models (L

R3D2: Realistic 3D Asset Insertion via Diffusion for Autonomous Driving Simulation

arXiv:2506.07826v2 Announce Type: replace Abstract: Validating autonomous driving (AD) systems requires diverse and safety-critical testing, making photorealistic virtual environments essential. Tradi

safetyarxiv-cs-cv

RAG-DIVE: A Dynamic Approach for Multi-Turn Dialogue Evaluation in Retrieval-Augmented Generation

arXiv:2604.16310v1 Announce Type: cross Abstract: Evaluating Retrieval-Augmented Generation (RAG) systems using static multi-turn datasets fails to capture the dynamic nature of real-world dialogues.

Industry

Really excited for this week! Next up, we've got something to show you at 12 pm PT today.

Sam Altman announced on X that OpenAI had something to reveal at 12 pm PT on a specific day, generating anticipation about an upcoming product announcement or demonstration. The post suggests a schedu

industrysam-altman--x

REFLEX: Self-Refining Explainable Fact-Checking via Verdict-Anchored Style Control

model-releasesarxiv-cs-cl

arXiv:2511.20233v3 Announce Type: replace Abstract: The prevalence of fake news on social media demands automated fact-checking systems to provide accurate verdicts with faithful explanations. However

Tutorials

RePrompT: Recurrent Prompt Tuning for Integrating Structured EHR Encoders with Large Language Models

arXiv:2604.17725v1 Announce Type: new Abstract: Large Language Models (LLMs) have shown strong promise for mining Electronic Health Records (EHRs) by reasoning over longitudinal clinical information t

tutorialsarxiv-cs-cl

Seeing Isn't Believing: Mitigating Belief Inertia via Active Intervention in Embodied Agents

arXiv:2604.17252v1 Announce Type: new Abstract: Recent advancements in large language models (LLMs) have enabled agents to tackle complex embodied tasks through environmental interaction. However, the

agentsarxiv-cs-cl

SFTMix: Elevating Language Model Instruction Tuning with Mixup Recipe

arXiv:2410.05248v4 Announce Type: replace Abstract: To acquire instruction-following capabilities, large language models (LLMs) undergo instruction tuning, where they are trained on instruction-respon

StageMem: Lifecycle-Managed Memory for Language Models

arXiv:2604.16774v1 Announce Type: new Abstract: Long-horizon language model systems increasingly rely on persistent memory, yet many current designs still treat memory primarily as a static store: wri

The Global Neural World Model: Spatially Grounded Discrete Topologies for Action-Conditioned Planning

arXiv:2604.16585v1 Announce Type: new Abstract: We present the Global Neural World Model (GNWM), a self-stabilizing framework that achieves topological quantization through balanced continuous entropy

agentsarxiv-cs-lg

Training Language Models to Use Prolog as a Tool

arXiv:2512.07407v2 Announce Type: replace Abstract: Language models frequently produce plausible yet incorrect reasoning traces that are difficult to verify. We investigate fine-tuning models to use P

UniDomain: Pretraining a Unified PDDL Domain from Real-World Demonstrations for Generalizable Robot Task Planning

arXiv:2507.21545v3 Announce Type: replace Abstract: Robotic task planning in real-world environments requires reasoning over implicit constraints from language and vision. While LLMs and VLMs offer st

applicationsarxiv-cs-ro

Untrained CNNs Match Backpropagation at V1: A Systematic RSA Comparison of Four Learning Rules Against Human fMRI

arXiv:2604.16875v1 Announce Type: new Abstract: A central question in computational neuroscience is whether the learning rule used to train a neural network determines how well its internal representa

safetyarxiv-cs-lg

Local Ai

VocabTailor: Dynamic Vocabulary Selection for Downstream Tasks in Small Language Models

arXiv:2508.15229v3 Announce Type: replace Abstract: Small Language Models (SLMs) provide computational advantages in resource-constrained environments, yet memory limitations remain a critical bottlen

local-aiarxiv-cs-cl

What If Consensus Lies? Selective-Complementary Reinforcement Learning at Test Time

arXiv:2603.19880v2 Announce Type: replace Abstract: Test-Time Reinforcement Learning (TTRL) enables Large Language Models (LLMs) to enhance reasoning capabilities on unlabeled test streams by deriving

researcharxiv-cs-lg

Who Watches the Watchmen? Humans Disagree With Translation Metrics on Unseen Domains

arXiv:2604.17393v1 Announce Type: new Abstract: Automatic evaluation metrics are central to the development of machine translation systems, yet their robustness under domain shift remains unclear. Mos

Why Agents Compromise Safety Under Pressure

arXiv:2603.14975v2 Announce Type: replace-cross Abstract: Large Language Model agents deployed in complex environments frequently encounter a conflict between maximizing goal achievement and adhering

Will People Enjoy a Robot Trainer? A Case Study with Snoopie the Pacerbot

arXiv:2604.18331v1 Announce Type: new Abstract: The physicality of exercise makes the role of athletic trainers unique. Their physical presence allows them to guide a student through a motion, demonst

agentsarxiv-cs-ro

a good reminder that your security is only as good as the weakest part of your security

This post likely emphasizes the 'weakest link' principle in cybersecurity, illustrating how comprehensive security systems can be compromised through a single vulnerable component, whether that's outd

agentsyohei-nakajima--x

A useful ward against slop story/science posts on X is noting which is in the character limit. All of the models struggle to do 280 characte…

applicationsethan-mollick--x

A useful ward against slop story/science posts on X is noting which is in the character limit. All of the models struggle to do 280 character summaries on their first pass, and most of the people crea

Local Ai

Adapting in the Dark: Efficient and Stable Test-Time Adaptation for Black-Box Models

arXiv:2604.15609v1 Announce Type: cross Abstract: Test-Time Adaptation (TTA) for black-box models accessible only via APIs remains a largely unexplored challenge. Existing approaches such as post-hoc

local-aiarxiv-cs-cv

AI Agents and Hard Choices

arXiv:2504.15304v2 Announce Type: replace Abstract: Can AI agents deal with hard choices -- cases where options are incommensurable because multiple objectives are pursued simultaneously? Adopting a t

safetyarxiv-cs-ai

Applied Explainability for Large Language Models: A Comparative Study

arXiv:2604.15371v1 Announce Type: cross Abstract: Large language models (LLMs) achieve strong performance across many natural language processing tasks, yet their decision processes remain difficult t

applicationsarxiv-cs-ai

AtManRL: Towards Faithful Reasoning via Differentiable Attention Saliency

model-releasesarxiv-cs-ai

arXiv:2604.16158v1 Announce Type: cross Abstract: Large language models (LLMs) increasingly rely on chain-of-thought (CoT) reasoning to solve complex tasks. Yet ensuring that the reasoning trace both

ATTNPO: Attention-Guided Process Supervision for Efficient Reasoning

arXiv:2602.09953v2 Announce Type: replace Abstract: Large reasoning models trained with reinforcement learning and verifiable rewards (RLVR) achieve strong performance on complex reasoning tasks, yet

Hardware

AutoFlows++: Hierarchical Message Flow Mining for System on Chip Designs

arXiv:2604.15359v1 Announce Type: cross Abstract: Understanding communication behavior in modern system-on-chip (SoC) designs is critical for functional verification, performance analysis, and post-si

hardwarearxiv-cs-lg

Beyond Passive Viewing: A Pilot Study of a Hybrid Learning Platform Augmenting Video Lectures with Conversational AI

arXiv:2604.15334v1 Announce Type: cross Abstract: The exponential growth of AI education has brought millions of learners to online platforms, yet this massive scale has simultaneously exposed critica

applicationsarxiv-cs-ai

C-Mining: Unsupervised Discovery of Seeds for Cultural Data Synthesis via Geometric Misalignment

arXiv:2604.15675v1 Announce Type: new Abstract: Achieving cultural alignment in Large Language Models (LLMs) increasingly depends on synthetic data generation. For such synthesis, the most vital initi

CiPO: Counterfactual Unlearning for Large Reasoning Models through Iterative Preference Optimization

arXiv:2604.15847v1 Announce Type: new Abstract: Machine unlearning has gained increasing attention in recent years, as a promising technique to selectively remove unwanted privacy or copyrighted infor

Consistency Analysis of Sentiment Predictions using Syntactic & Semantic Context Assessment Summarization (SSAS)

model-releasesarxiv-cs-ai

arXiv:2604.15547v1 Announce Type: cross Abstract: The fundamental challenge of using Large Language Models (LLMs) for reliable, enterprise-grade analytics, such as sentiment prediction, is the conflic

Cut Your Losses! Learning to Prune Paths Early for Efficient Parallel Reasoning

arXiv:2604.16029v1 Announce Type: new Abstract: Parallel reasoning enhances Large Reasoning Models (LRMs) but incurs prohibitive costs due to futile paths caused by early errors. To mitigate this, pat

DeepER-Med: Advancing Deep Evidence-Based Research in Medicine Through Agentic AI

arXiv:2604.15456v1 Announce Type: new Abstract: Trustworthiness and transparency are essential for the clinical adoption of artificial intelligence (AI) in healthcare and biomedical research. Recent d

agentsarxiv-cs-ai

Detecting and Suppressing Reward Hacking with Gradient Fingerprints

arXiv:2604.16242v1 Announce Type: cross Abstract: Reinforcement learning with verifiable rewards (RLVR) typically optimizes for outcome rewards without imposing constraints on intermediate reasoning.

Eco-Bee: A Personalised Multi-Modal Agent for Advancing Student Climate Awareness and Sustainable Behaviour in Campus Ecosystems

arXiv:2604.15327v1 Announce Type: cross Abstract: Universities are microcosms of urban ecosystems, with concentrated consumption patterns in food, transport, energy, and product usage. These environme

agentsarxiv-cs-ai

FACTS: Table Summarization via Offline Template Generation with Agentic Workflows

arXiv:2510.13920v2 Announce Type: replace Abstract: Query-focused table summarization requires generating natural language summaries of tabular data conditioned on a user query, enabling users to acce

agentsarxiv-cs-cl

FSPO: Few-Shot Optimization of Synthetic Preferences Personalizes to Real Users

arXiv:2502.19312v2 Announce Type: replace-cross Abstract: Effective personalization of LLMs is critical for a broad range of user-interfacing applications such as virtual assistants and content curati

applicationsarxiv-cs-ai

How people use Copilot for Health

arXiv:2604.15331v1 Announce Type: cross Abstract: We analyze over 500,000 de-identified health-related conversations with Microsoft Copilot from January 2026 to characterize what people ask conversati

safetyarxiv-cs-ai

Improving Reasoning Capabilities in Small Models through Mixture-of-Layers Distillation with Stepwise Attention on Key Information

arXiv:2604.15701v1 Announce Type: new Abstract: The significant computational demands of large language models have increased interest in distilling reasoning abilities into smaller models via Chain-o