Search: “llm”

RingSQL: Schema-Independent Synthetic Data Generation for Text-to-SQL Reinforcement Learning

arXiv:2601.05451v2 Announce Type: replace-cross Abstract: Recent advances in text-to-SQL have been driven by larger models, better datasets, and new training methods like RLVR. However, progress remai

6 Aug 2026

Content type

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

Research

Spend Bits Where Queries Look: KV Cache Vector Quantization with Attention-Preserving Transforms

DGX agent

arXiv:2608.04074v1 Announce Type: cross Abstract: Long-context LLM decoding reads the key-value (KV) cache at every step. Loading it takes longer than computing attention over it, so throughput is ban

researcharxiv-cs-ai

6 Aug 2026

Agentic Reinforcement Learning with Self-Distilled Reward Shaping

arXiv:2608.03223v1 Announce Type: cross Abstract: Agentic reinforcement learning enables LLM agents to learn through interaction, but sparse trajectory-level rewards reveal success without identifying

AnchorKV: Anchor-Residual KV Cache Compression

arXiv:2608.02901v1 Announce Type: cross Abstract: The key-value (KV) cache is the primary memory bottleneck in long-context LLM inference. Existing approaches attack it from opposite ends: eviction me

Anyone interested in building a harness-only benchmark?

model-releasesr-localllama

There are a lot of LLM benchmarks but few, if any, harness benchmarks. I am thinking this would be a really good community project to build one. End goal: a leaderboard of harness performance (multipl

ARCHead: Activation-Metric Residual Correction for Large Language Model Output Heads

arXiv:2608.02703v1 Announce Type: new Abstract: Weight-only quantization substantially reduces the storage of large language model (LLM) transformer blocks, but practical backends often retain the fin

Autoreflection: How Agentic Strange Loops Turn Human Culture into AI Infrastructure

arXiv:2608.03800v1 Announce Type: cross Abstract: An LLM-based agent is a loop that reads itself. Agentic frameworks externalize identity, memory, and disposition into editable files. The agent loads

Benchmarking the Benchmarks: Testing the Predictive Validity of Commonsense Benchmarks

arXiv:2608.03340v1 Announce Type: new Abstract: Predicting LLM's capabilities on real-world tasks is essential, yet the extent to which performance on commonsense benchmarks predicts downstream perfor

applicationsarxiv-cs-cl

Formal Verification of Agentic Systems over Operational Data

arXiv:2608.03609v1 Announce Type: new Abstract: Agentic systems driven by large language models (LLMs) are increasingly deployed in real-world workflows where they act on persistent operational data.

Local Ai

Interpreting Black-Box Large Language Models with Sentence-Level Energy Landscapes

arXiv:2608.02879v1 Announce Type: new Abstract: The widespread adoption of proprietary Large Language Models (LLMs) accessed strictly through closed APIs has created a critical challenge for responsib

local-aiarxiv-cs-ai

ISEE: Interactive Semantic Enrichment for Database Fields

arXiv:2608.02604v1 Announce Type: new Abstract: LLM-based agents are increasingly being deployed for data-related tasks, including data sense-making, exploration, and retrieval. However, their perform

applicationsarxiv-cs-ai

On the missing benchmarks layer and a potential solution

arXiv:2608.02996v1 Announce Type: new Abstract: Latin America is missing a foundational layer for native AI development: the benchmark layer. The benchmark layer does two things no other layer can - i

OncoTriad-QA: A Patient-Level Radiology-Pathology-Genomics Benchmark for Pan-Cancer Reasoning

arXiv:2608.02615v1 Announce Type: cross Abstract: Cancer diagnosis and characterization require integrating complementary evidence from radiology, pathology, genomics, and clinical metadata. However,

OR-Agent: Bridging Evolutionary Search and Structured Research for Automated Algorithm Discovery

arXiv:2602.13769v3 Announce Type: replace Abstract: Automating heuristic design in complex, experiment-driven domains requires more than iterative mutation of solution algorithms. Current LLM-based ev

Preferred, Not Safer: Pairwise Preference Is a Poor Proxy for Clinical Safety

arXiv:2608.02617v1 Announce Type: cross Abstract: We evaluate whether clinician pairwise preferences provide a reliable signal of clinical safety in large language model (LLM) evaluation using expert

S^3: Improving Agent Safety through Multi-Stage Defense

arXiv:2608.02683v1 Announce Type: cross Abstract: Large Language Model (LLM) agents rely on multi-stage agentic workflows, with stages such as memory, planning, and tool execution, to accomplish compl

ToolLIFT: Lifting Tool-Specific Trajectories into Function-Level Graphs for Generalizable Tool Planning

arXiv:2608.03468v1 Announce Type: new Abstract: Historical tool-use trajectories provide valuable experience for large language model (LLM) agents to plan and coordinate tool usage. Existing approache

AgentMemBench: A Systematic Benchmark for Evaluating Long-Term Memory Management Strategies in Conversational AI Agents

arXiv:2608.00009v1 Announce Type: new Abstract: Long-term memory remains a critical bottleneck for conversational AI agents, whose finite context windows cannot support coherent recall across thousand

Hardware

Bole: Efficient Tree Speculation for Hybrid-Attention Language Models

arXiv:2608.01651v1 Announce Type: cross Abstract: Hybrid-attention large language models combine full attention with recurrent linear attention to reduce long-context inference costs, yet their autore

hardwarearxiv-cs-cl

CompanionBench: A Theory-Anchored, Real-World-Grounded Benchmark for AI Emotional Companionship

arXiv:2608.02046v1 Announce Type: new Abstract: LLM companions are deployed at scale in personally consequential settings, yet poorly evaluated. Existing benchmarks use hand-authored scenarios and pro

DeBERTa-Sentinel: Toward Transparent and Trustworthy Detection of AI-Generated Text

arXiv:2608.01046v1 Announce Type: new Abstract: The rapid spread of large language models (LLMs) across the web raises concerns about misinformation, academic integrity, automated content manipulation

Enhancing Large Language Model Reasoning with Reward Models: An Analytical Survey

arXiv:2510.01925v3 Announce Type: replace Abstract: Reward models (RMs) play a critical role in enhancing the reasoning performance of LLMs. For example, they can provide training signals to finetune

From Chains to Trees: Parent-Conditioned Drafting for Semi-Autoregressive Speculative Decoding

arXiv:2608.02123v1 Announce Type: new Abstract: Speculative decoding accelerates LLM inference only when drafted continuations survive target-model verification. Semi-autoregressive drafters such as D

I benchmarked the 4 models I had pulled. The 1.1GB one beat the 2GB one at math and lost badly at extraction.

152 generations, deterministic grading (exact number/string/JSON/regex), no LLM judge on my 16GB laptop. task type | deepseek-r1:1.5b (1.1GB) | llama3.2:3b (2.0GB) | gemma:2b | codellama 7b arithmetic

model-releasesr-ollama

LongHorizon-Harness: Advancing Long-Horizon Agents for Real-World Tasks

model-releasesarxiv-cs-cv

arXiv:2608.01964v1 Announce Type: new Abstract: Large language model (LLM) agents increasingly undertake long-horizon tasks that require sustained reasoning, tool use, and revision across many interde

MedUPS: Towards Diagnostic Assistance in Uncommon Medical Cases with Large Language Models

arXiv:2608.01012v1 Announce Type: new Abstract: Uncommon and off-guideline cases are difficult for clinical decision support, because physicians must make a series of management decisions under diagno

safetyarxiv-cs-cl

No One Wins in Nuclear War: A Social Simulation of Military Decision-making

arXiv:2608.01868v1 Announce Type: cross Abstract: WOPR is a social-simulation environment for studying how organizations make high-stakes decisions, built on a deterministic, replay-validated rules en

safetyarxiv-cs-cl

SLMs as Multi-Agent Routers: A Progressive SFT and Reinforcement Learning Approach

arXiv:2608.00030v1 Announce Type: new Abstract: Specialised retrieval agents typically surface higher quality results than general-purpose search, but selecting the optimal agent for a given query rem

Local Ai

TRACE-TS: Attribution-Grounded and Traceable Sensor-Language Reasoning for Human Activity Understanding

arXiv:2608.00200v1 Announce Type: cross Abstract: Wearable sensors capture fine-grained motion patterns that support rich behavioral understanding, yet most existing methods reduce these signals to ac

local-aiarxiv-cs-cl

Unpacking Hateful Memes: Presupposed Context and False Claims

arXiv:2510.09935v2 Announce Type: replace Abstract: While memes are often humorous, they are frequently used to disseminate hate, causing serious harm to individuals and society. Current approaches to

Visualising Information Flow in Word Embeddings with Diffusion Tensor Imaging

arXiv:2601.05713v2 Announce Type: replace Abstract: Understanding how large language models (LLMs) represent natural language is a central challenge in natural language processing (NLP) research. Many

When Prompts Control Robots: Prompt Injection Attacks in Multi-Agent Robotic Systems

arXiv:2608.00747v1 Announce Type: new Abstract: Large language models are increasingly integrated into autonomous robotic systems for task planning and control, but this integration exposes them to pr

agentsarxiv-cs-ro

When Replanning Becomes the Bottleneck: Budgeted Replanning for Embodied Agents

arXiv:2608.01428v1 Announce Type: cross Abstract: Embodied agents replan frequently to recover from execution drift, partial observability, and coordination hazards, but each LLM-based replanning call

researcharxiv-cs-lg

Autonomous Repair for Multi-Agent Systems via Monte-Carlo Tree Search

arXiv:2607.29055v1 Announce Type: cross Abstract: Multi-agent systems (MAS) are increasingly deployed to solve complex tasks. In case of incorrect or unsatisfactory outputs, users have to manually loc

CAGE: Certified Authorization under Typed-Return Uncertainty for Tool-Using Agents

arXiv:2607.29190v1 Announce Type: new Abstract: Tool-using LLM agents act on typed tool returns, records pairing provenance and categorical fields with numerical values. Runtime permission gates gener

Can Large Language Models Derive New Knowledge? A Dynamic Benchmark for Biological Knowledge Discovery

arXiv:2603.03322v2 Announce Type: replace-cross Abstract: Recent advancements in Large Language Model (LLM) agents have demonstrated remarkable potential in automatic knowledge discovery. However, rig

Local Ai

Deconstructing Off-Policy Ratios: Entropy-Scaled Trust Regions for Asynchronous Reinforcement Learning

arXiv:2607.22186v2 Announce Type: replace Abstract: Asynchronous reinforcement learning (RL) accelerates large language model (LLM) post-training by overlapping rollout generation with policy optimiza

local-aiarxiv-cs-ai

Distilling Knowledge from Large Language Models into Lightweight Reinforcement Learning Agents for Autonomous Cyber Operations

model-releasesarxiv-cs-lg

arXiv:2607.28826v1 Announce Type: new Abstract: Autonomous Cyber Operations (ACO) are increasingly important for defending enterprise networks as cyber threats continue to evolve in sophistication. AC

RePaCA: Leveraging Reasoning Large Language Models for Static Automated Patch Correctness Assessment

arXiv:2507.22580v2 Announce Type: replace-cross Abstract: Automated Program Repair (APR) seeks to automatically correct software bugs without requiring human intervention. However, existing tools tend

Semantics of Subterfuge: Benchmarking Legal Deception Detection Against General-domain State-of-the-Art

arXiv:2607.29066v1 Announce Type: cross Abstract: Deception detection has critical implications for legal proceedings, law enforcement, and online security. Although human judgment is limited in accur

applicationsarxiv-cs-ai

DeepSeek V4 Flash 0731 IQ2_M benchmark for Dual 3060 and 96GB RAM ≈ 3.5 tok/s.

model-releasesr-localllama

Thanks to the community help I finally launched this llm. LM Studio refused to load weight onto second GPU but Unsloth Studio did so everything was done in there. Not a proper benchmark (used PC in pa

1 Aug 2026

AI LEGO: Scaffolding Cross-Functional Collaboration in Industrial Responsible AI Practices during Early Design Stages

arXiv:2505.10300v2 Announce Type: replace-cross Abstract: Responsible AI (RAI) efforts increasingly emphasize the importance of addressing potential harms early in the AI development lifecycle through

AI systems and the reproduction of (standard) language ideologies in World Englishes

arXiv:2607.28528v1 Announce Type: new Abstract: The rapid growth of large language models (LLMs) has resurrected age-old questions in sociolinguistics and world Englishes, such as who decides what cou

applicationsarxiv-cs-cl

Benchmarking the Residual: What Long-Horizon Evaluations Add Beyond Matched Short-Task Performance

arXiv:2607.27283v1 Announce Type: new Abstract: Long-horizon benchmarks often show that agents fail more as tasks become longer. This observation is useful for deployment, but it does not by itself ex

agentsarxiv-cs-lg

Beyond a Single Judge: Simulating Social Persona Panels for Generative UI Evaluation

arXiv:2607.28439v1 Announce Type: new Abstract: Generative UI (GenUI) lets large language models synthesize a complete, renderable interface directly from a natural-language instruction, but evaluatin

Beyond Borrowed Histories: Person-Aligned User Simulation for Interactive Role-Playing Evaluation

arXiv:2607.27816v1 Announce Type: new Abstract: Role-playing agents (RPAs) have become one of the most important consumer applications of large language models. Users engage in multi-turn conversation

Beyond KV Reconstruction: Functional Reconstruction for MLA Draft Models in Speculative Decoding

model-releasesarxiv-cs-lg

arXiv:2607.27269v1 Announce Type: new Abstract: Multi-head latent attention (MLA) is increasingly important for long-context LLM inference because compact latent states replace the growing key-value (

Can Agents Deceive? Evaluating Reasoning and Deception in ParliamentBench using a Social Deduction Game

arXiv:2607.28146v1 Announce Type: new Abstract: As large language models (LLMs) are deployed as agents in high-stakes settings, such as medical and legal systems, understanding their deceptive capabil