Search: “benchmark”

One Policy, Any Budget: Internalizing Budget-Aware Search via Reinforcement Learning

arXiv:2609.00813v1 Announce Type: new Abstract: While reinforcement learning has enabled LLM-based search agents to invoke external tools, existing methods train under fixed budgets and cannot adapt w

Online Self-Weighted Fine-Tuning

Content type

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

DGX agent

arXiv:2609.00734v1 Announce Type: new Abstract: Standard supervised fine-tuning (SFT) assigns the same explicit loss weight to every expert demonstration, regardless of the model's changing competence

researcharxiv-cs-lg

2 Sept 2026

Ontology-Guided Neuro-Symbolic Inference: Grounding Language Models with Mathematical Domain Knowledge

arXiv:2602.17826v2 Announce Type: replace Abstract: Language models exhibit fundamental limitations -- hallucination, brittleness, and lack of formal grounding -- that are particularly problematic in

Patterning in Practice: Debiasing Reward Models with Susceptibilities

arXiv:2609.00699v1 Announce Type: new Abstract: Reward models trained on human preferences are known to suffer from length, formatting, and other stylistic biases. In this paper we use patterning, whi

PersuaRL: Reinforcement Learning-Driven Multi-Expert Selection for Persuasive Dialogue Generation in Insurance

model-releasesarxiv-cs-cl

arXiv:2609.01188v1 Announce Type: new Abstract: Large Language Models (LLMs) are revolutionizing digital communication by powering conversational agents deployed across domains such as customer servic

Polarizable atomic multipoles for learning long-range electrostatics

arXiv:2605.05746v2 Announce Type: replace-cross Abstract: Long-range electrostatics and polarization remain central obstacles to extending machine learning interatomic potentials (MLIPs) to ionic, pol

researcharxiv-cs-lg

Agents

Polished but Unresolved: Identifying Late-Stage Pressure States in Long-Horizon Tool-Use Agents

arXiv:2609.00823v1 Announce Type: new Abstract: Long-horizon tool-use agents need not only to search and plan, but also to decide when to finalize. We study late-stage pressure states, in which an age

agentsarxiv-cs-ai

PopPert: Population-level Joint-Distribution Modeling for Single-Cell Perturbation Prediction

arXiv:2609.01357v1 Announce Type: cross Abstract: Predicting transcriptional responses to specific perturbations is critical for understanding cellular regulatory mechanisms and accelerating drug disc

Potential-Guided Particle Steering for Negation-Constrained Dexterous Grasping

model-releasesarxiv-cs-cv

arXiv:2609.00555v1 Announce Type: cross Abstract: Language-driven dexterous grasp models, such as DextER, perform well when instructions specify where to grasp, but we find they fail systematically wh

Prediction-Assisted Pricing and Admission for LLM APIs with Stochastic Token Consumption

arXiv:2609.00710v1 Announce Type: cross Abstract: An LLM application often sells or internally allocates several service products: a small or premium model, a short or long token cap, and possibly mul

Process-Aware AI for Rainfall-Runoff Modeling: A Mass-Conserving Neural Framework with Hydrological Process Constraints

arXiv:2603.25093v2 Announce Type: replace Abstract: Machine learning models can achieve high predictive accuracy in hydrological applications but often lack physical interpretability. The Mass-Conserv

Quantum Sparse Autoencoders for Q-Matrix Estimation in Cognitive Diagnosis

arXiv:2609.01537v1 Announce Type: new Abstract: Q-matrices play a central role in cognitive diagnosis within educational data mining (EDM), specifying which latent skills each assessment item requires

Qwen3.8-flash-next sees corruption everywhere

model-releasesr-localllama

Hi, I've noticed that the model often sees 'garbled text' in its context. Sometimes it declare that the tools instructions are corrupted, sometimes it is the content of some .md files, ora other files

RadMatch: Auditable Radiology Report Evaluation via Finding-Level Matching

arXiv:2609.01470v1 Announce Type: new Abstract: As AI systems are increasingly used to draft radiology reports, reliably evaluating their clinical quality remains a critical challenge. Large language

safetyarxiv-cs-cv

Real-Time Neuromorphic Spectrum Intelligence Simulator

arXiv:2609.00585v1 Announce Type: cross Abstract: We present the Real-Time Neuromorphic Spectrum Intelligence Simulator (RT-NuSIS), a modular framework to study spiking neural network (SNN) and memris

ReDeck: Step-Level Render-Grounded Refinement for Document-to-Slide Generation

arXiv:2609.00194v1 Announce Type: new Abstract: Document-to-slide generation is challenging because slides are dense editable artifacts that require both faithful content selection and precise spatial

Relational Task Generation Language: A Declarative Specification Framework for Relational Deep Learning

arXiv:2609.01292v1 Announce Type: cross Abstract: Relational Deep Learning (RDL) has become a powerful paradigm for learning from multi-tabular data. However, manually defining RDL prediction tasks is

Reliability Challenges in Diffusion Vision-Language Models

arXiv:2609.01318v1 Announce Type: cross Abstract: Diffusion-based Large Vision-Language Models (dLVLMs) have recently emerged as a compelling alternative to autoregressive (AR) LVLMs, offering advanta

safetyarxiv-cs-cl

Replacing Training with Memory: Listwise Selection for Text-to-SQL

arXiv:2609.00834v1 Announce Type: cross Abstract: Modern Text-to-SQL systems often follow generate-execute-select pipelines, generating multiple candidate queries then selecting the best one. Listwise

Retrieved but not ranked: surface-form bias in structural retrieval, from mathematics to agent trajectories

arXiv:2609.01556v1 Announce Type: cross Abstract: We evaluate embedding retrieval where surface form and meaning are pulled apart on purpose: retrieving items that share underlying structure but not w

SARTM: Segment Any RGB Thermal Model with Language aided Distillation

arXiv:2505.01950v2 Announce Type: replace-cross Abstract: The recent Segment Anything Model (SAM) demonstrates strong instance segmentation performance across various downstream tasks. However, SAM is

Scene Graph-based Driving Scenario Extraction for Automotive Egocentric Datasets

model-releasesarxiv-cs-ro

arXiv:2609.00333v1 Announce Type: new Abstract: Extracting scenarios from unlabelled real-world sensor data streams is a critical but challenging task in the development process of automated driving s

SCoNE: Selective Context-aware Neuron Editing for Robust Retrieval-Augmented Generation

arXiv:2609.00689v1 Announce Type: new Abstract: Retrieval-Augmented Generation (RAG) is highly sensitive to retrieval noise: when retrieved documents mix informative and irrelevant context, LLMs are e

researcharxiv-cs-cl

Applications

SEBA: Sample-Efficient Black-Box Attacks on Visual Reinforcement Learning

arXiv:2511.09681v2 Announce Type: replace-cross Abstract: Visual reinforcement learning has achieved remarkable progress in visual control and robotics, but its vulnerability to adversarial perturbati

applicationsarxiv-cs-ai

Semi-Supervised Biomedical Image Segmentation via Diffusion Models and Teacher-Student Co-Training

model-releasesarxiv-cs-cv

arXiv:2504.01547v3 Announce Type: replace Abstract: Supervised deep learning achieves strong performance in biomedical image segmentation but relies on costly pixel-wise annotations, motivating semi-s

Silence is Golden: Mitigating Hallucinations in Large Audio-Language Models via Layer-Weighted Vector Steering

arXiv:2510.12851v2 Announce Type: replace-cross Abstract: Large Audio-Language Models (LALMs) excel in Audio QA but often suffer from hallucinations ungrounded in the audio. To our knowledge, we are t

SinkPruner: Sink-Free Visual Token Pruning for Multimodal Large Language Models

arXiv:2609.01004v1 Announce Type: cross Abstract: Despite their strong multimodal understanding ability, multimodal large language models (MLLMs) incur substantial computational overhead when processi

SMELT: Scaling Laws for Compute-Matched MoE Looped Transformers

arXiv:2609.01343v1 Announce Type: new Abstract: Looped Transformers increase effective depth by iterating a shared block of layers, but most evaluations compare at fixed model size, conflating archite

safetyarxiv-cs-lg

Applications

SO excited to share my next chapter: i’m joining @llama_index to lead all things growth marketing, brand, and community! 🦙🩷 when i came al…

SO excited to share my next chapter: i’m joining @llama_index to lead all things growth marketing, brand, and community! 🦙🩷 when i came along them, it clicked incredibly fast: a team building the solu

applicationsjerry-liu--x

SoK: When Safe Agents Fail Together: The Security of Multi Agent LLM Systems

arXiv:2609.00595v1 Announce Type: cross Abstract: Safe agents can fail together. Multi-agent LLM systems (MAS) move information, state, decisions, and authority across principal boundaries, creating f

StreamScout: Learning When to Look Deeper for Streaming Video Understanding

arXiv:2609.00291v1 Announce Type: new Abstract: Streaming video understanding requires answering questions that arrive at arbitrary moments over an unbounded video stream. Existing systems primarily f

safetyarxiv-cs-cv

Tutorials

SymFold: Synergizing Evolutionary and Structural Priors for Accurate Protein Inverse Folding

arXiv:2609.01353v1 Announce Type: new Abstract: Protein inverse folding aims to recover amino acid sequences for a given 3D protein structure, underpinning broad applications such as enzyme engineerin

tutorialsarxiv-cs-ai

Synthetic Worlds for Temporal Evaluation and Knowledge Updating in LLMs

model-releasesarxiv-cs-cl

arXiv:2609.00184v1 Announce Type: new Abstract: Large language models (LLMs) rely on static pretraining corpora, causing their knowledge to become outdated over time. Existing approaches for evaluatin

Task-Specific Prompt with Global Context for Multi-Task Graph Pre-Training

arXiv:2609.00047v1 Announce Type: cross Abstract: Graph prompt learning is an effective paradigm to adapt pre-trained graph models to downstream tasks in low-resource scenarios. However, existing mult

Agents

TEIDAN: A Multilingual Multiparty Dialogue Corpus

arXiv:2609.00802v1 Announce Type: new Abstract: Multi-party interaction is a central setting for human communication and a necessary target for human-agent interaction systems that must participate in

agentsarxiv-cs-cl

Text-guided flow matching enables sample-efficient crystal structure generation

arXiv:2609.01076v1 Announce Type: cross Abstract: Crystal generators can now propose periodic structures, but their control interfaces remain poorly matched to the mixed descriptors used in materials

The Curse of Multilinguality in Lexical Normalization

arXiv:2609.00329v1 Announce Type: cross Abstract: Lexical normalization rewrites the noisy, non-standard words that fill user-generated text (tmrw, u, gr8) into their standard forms. Because labelled

TimeSteer: Inference-Time Speech Scheduling in Joint Audio-Visual Diffusion Models

arXiv:2609.01277v1 Announce Type: cross Abstract: Although pretrained joint audio-visual diffusion models offer rich control over what to generate, they provide no explicit control over when an uttera

Towards Effective Structured Context Modeling for Conversational Recommender Systems via Dual-node Monte Carlo Tree Search

arXiv:2609.00618v1 Announce Type: cross Abstract: We investigate the role of conversational context modeling in user preference tracking for Conversational Recommendation Systems (CRSs). In this regar

Training-Free Policy Violation Detection via Activation-Space Whitening in LLMs

arXiv:2512.03994v4 Announce Type: replace Abstract: As organizations increasingly deploy LLMs in sensitive domains such as legal, financial, and medical settings, ensuring alignment with internal orga

safetyarxiv-cs-lg

Triple-Bottom-Line Sustainability of Language Models for Edge AI: A Comparison Between SLMs and Quantized LLMs

arXiv:2609.00665v1 Announce Type: new Abstract: Edge-AI model selection is commonly driven by one isolated metric - accuracy, latency, memory, energy, or safety, even though a deployable language mode

Trust Your Guide Only When Certain: Uncertainty-Aware Sparse Alignment at Inference Time

arXiv:2609.00624v1 Announce Type: new Abstract: A prominent paradigm in inference-time alignment employs lightweight supervisors to steer Large Language Models (LLMs). Through empirical analysis, we i

safetyarxiv-cs-cl

TWIX: a Two-Stage Approach for End-To-End Named Entity Recognition and Relation Extraction

model-releasesarxiv-cs-cl

arXiv:2609.00832v1 Announce Type: new Abstract: The exponential growth of scientific publications calls for automatic Information Extraction (IE) systems to support knowledge discovery. In this contex

Validity-Aware Jailbreak Evaluation for Large Language Models

arXiv:2609.00498v1 Announce Type: new Abstract: Jailbreak robustness has become central to large language model (LLM) safety evaluation, yet prevailing methodologies rely primarily on refusal behavior

VerNav: Verifier-First Low-Latency Vision-and-Language Navigation

model-releasesarxiv-cs-cv

arXiv:2609.00920v1 Announce Type: cross Abstract: Vision-and-Language Navigation (VLN) requires an agent to navigate through unseen 3D environments according to natural-language instructions. Explicit

ViTAMINS: An Empirical Study of Training Self-Supervised Vision Transformers with Synthetic Hard Negatives

arXiv:2609.01041v1 Announce Type: cross Abstract: We introduce ViTAMINS, a method that integrates synthetic hard negatives into unsupervised vision transformer pretraining to improve representation qu

When Features Become Instances: Inverted Contrastive Learning for Unsupervised Feature Selection

arXiv:2609.00782v1 Announce Type: new Abstract: Unsupervised feature selection seeks a compact subset of informative features without access to class labels, making feature utility difficult to define

Your favorite fastest abliterated/safety removed 3.6 and 3.8 27b?

model-releasesr-localllama

Not written by AI all mistakes mine. I saw people on the subreddit saying that 3.6 works better without thinking. It made me want to know for certain about which is better, 3.6 or 3.8 for low thinking