Search: “llm”

I-CALM: Incentivizing Confidence-Aware Abstention for LLM Selective Answering

arXiv:2604.03904v2 Announce Type: replace-cross Abstract: Large language models (LLMs) often produce confident but incorrect answers, in part because standard evaluation incentives reward guessing ove

Content type

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

Le Critique: Privileged Value Functions for LLM Reinforcement Learning

DGX agent

arXiv:2608.16739v1 Announce Type: new Abstract: Reinforcement learning algorithms for Large Language Models (LLMs) are largely distinguished by their variance reduction strategy. Group-relative method

safetyarxiv-cs-lg

18 Aug 2026

Mental Model Management: An Operator-Based Framework for LLM Memory

arXiv:2608.15451v1 Announce Type: new Abstract: Large language models process large amounts of information but usually lack an explicit mechanism for maintaining compact and evolving conceptual repres

Agents

MUSE: An Interactive Meta-Agent for Understanding and Steering LLM-powered Data Science Systems

arXiv:2608.16181v1 Announce Type: cross Abstract: Recent advances in large language models have enabled a new class of agentic data science systems that allow users to complete complex data science wo

agentsarxiv-cs-ai

OTel: Building Domain-Specialized Telecom LLM Foundations for Intelligent Networks

arXiv:2608.15436v1 Announce Type: new Abstract: Frontier AI models have advanced rapidly, but they still struggle with telecom-specific tasks. We present Open Telco (OTel), an open telecom AI resource

PCA-guided Activation Scaling for Monotonic Bidirectional Control over LLM Sycophancy

arXiv:2608.16650v1 Announce Type: new Abstract: Large language models (LLMs) exhibit sycophancy, a tendency to agree with user beliefs regardless of factual accuracy. This can reinforce misconceptions

researcharxiv-cs-cl

PertMind: Eliciting Emergent Biological Reasoning in LLM via Reinforcement Learning on Cellular Perturbation Data

arXiv:2608.16419v1 Announce Type: cross Abstract: Large language models can describe mechanisms, yet scalable post-training still depends on costly, manually curated biological reasoning traces. Here

Prior Audit-Repair Context Shifts LLM Verifier Thresholds Toward Leniency

arXiv:2608.16003v1 Announce Type: new Abstract: Automated checking pipelines increasingly place one language model as the checker and another (or the same one) as the fixer. We ask whether that wiring

SchurQuant: Groupwise Discrete Optimization for Layer-Wise LLM Quantization

model-releasesarxiv-cs-lg

arXiv:2608.15567v1 Announce Type: new Abstract: Weight-only post-training quantization (PTQ) enables the deployment of large language models under tight memory budgets, but accuracy often collapses at

STAGE: Controlled Objective Admission for Multi-Preference LLM Alignment

model-releasesarxiv-cs-cl

arXiv:2608.16553v1 Announce Type: new Abstract: Multi-preference alignment is often framed as scalarization: combine reward dimensions, then optimize. This leaves a temporal decision underspecified: w

SubZero+: Efficient Zeroth-Order LLM Fine-Tuning via Large Learning Rates

model-releasesarxiv-cs-lg

arXiv:2608.15665v1 Announce Type: new Abstract: Zeroth-order (ZO) optimization enables backpropagation-free fine-tuning of large language models, but existing ZO methods suffer from high-variance grad

Value Leakage: An LLM's Answers Are Silently Shaped by Its Own Values

arXiv:2607.14345v4 Announce Type: replace-cross Abstract: People use language models for practical questions whose answers are difficult to verify. We show that models exhibit covert value leakage: th

Ventor-QTest: Threat-Model-Driven Verification of Vendor-Hosted LLM APIs

arXiv:2608.16391v1 Announce Type: cross Abstract: As large language models become increasingly widespread, third-party providers that deploy open-weight models have become an important part of the eco

WeSCE: A Benchmark for Measuring Security Drift in LLM-Driven Code Editing

arXiv:2608.15092v1 Announce Type: cross Abstract: In this work, we introduce WeSCE, a benchmark for quantifying security drift in code editing under weak-security constraints, where tasks specify only

Applications

Musical Mirrors: The LLM as Sounding Board in Songwriting

arXiv:2608.13944v1 Announce Type: cross Abstract: This paper examines a use of AI in creative practice as an interpretive sounding board for human-generated material, rather than the more familiar pat

applicationsarxiv-cs-ai

17 Aug 2026

Tutorials

No Universal Signal Predicts Sample-Level LLM Regression under Version Updates

arXiv:2608.13607v1 Announce Type: new Abstract: Frontier LLMs are updated frequently and typically outperform their predecessors in aggregate. But aggregate gains say little about individual samples:

tutorialsarxiv-cs-ai

17 Aug 2026

SemPlan: Benchmarking Structured Semantic Planning for LLM-Based Queries over Enterprise Data

arXiv:2608.13612v1 Announce Type: new Abstract: Natural-language interfaces to enterprise data must translate underspecified requests into governed, executable behavior while controlling invalid queri

17 Aug 2026

Beyond the Best Guess: Improving LLM Solution Coverage with Evolution Strategies

arXiv:2608.12679v1 Announce Type: new Abstract: Large Language Models (LLMs) are increasingly deployed in discovery domains such as math and science. The usual approach is to present the problem to th

From Caveman to Expert Analyst: Energy Consumption of Variable LLM Tasks

arXiv:2608.12350v1 Announce Type: cross Abstract: The energy demand growth and environmental impacts of artificial intelligence (AI) have generated substantial interest in supplying sufficient low-cos

Hardware

Reduced Matrix Multiplication: Input-Adaptive Matrix-Product Reduction for LLM Inference

arXiv:2608.13426v1 Announce Type: cross Abstract: Transformer-based language models achieve strong performance but incur substantial inference cost due to repeated high-dimensional matrix multiplicati

hardwarearxiv-cs-ai

SkillShapley: Boundary-Adaptive Shapley Valuation for Skill Step Attribution in LLM Agents

arXiv:2608.13173v1 Announce Type: new Abstract: Agent skills are crucial external instructions that enable language agents to execute long procedural tasks such as coding or document processing. Exist

Teach the Magnitude, Not the Direction: Verifier-Bounded Credit Assignment for Multi-Turn Multi-step LLM Agents

arXiv:2608.13179v1 Announce Type: new Abstract: Reinforcement learning with verifiable rewards (RLVR) offers a verifier-bounded performance ceiling for training multi-turn tool-use agents, yet its tra

A-3PO: Accelerating Asynchronous LLM Training with Staleness-aware Proximal Policy Approximation

arXiv:2512.06547v4 Announce Type: replace-cross Abstract: Decoupled PPO has been a successful reinforcement learning (RL) algorithm to deal with the high data staleness under the asynchronous RL setti

Agents

Credo: Declarative Control of LLM Pipelines via Beliefs and Policies

arXiv:2604.14401v2 Announce Type: replace Abstract: Agentic AI systems are becoming commonplace in domains that require long-lived, stateful decision-making in continuously evolving conditions. As suc

agentsarxiv-cs-ai

Hardware

Cutting AI Datacenter Energy with Reinforcement Learning: Measured Power Control of LLM Training from One GPU to the Fleet

arXiv:2608.11226v1 Announce Type: new Abstract: Reinforcement-learning post-training dominates modern language-model development, yet its power behavior on GPU hardware has not been characterized, and

hardwarearxiv-cs-ai

EnterpriseRAG: Benchmarking LLM Instruction Adherence and Robustness under Non-Ideal Enterprise Retrieval

arXiv:2608.11584v1 Announce Type: new Abstract: Enterprise RAG deployments face a critical reliability gap: while LLMs satisfy 80% of individual constraints, only 26.8% of responses meet all requireme

From Prompting to Behavioral Alignment: Personalized LLM Judges for Recommendation Evaluation

arXiv:2608.11493v1 Announce Type: new Abstract: Traditional offline recommendation evaluation relies heavily on complex, manually maintained feature pipelines that are difficult to scale. While Large

Graph-Structured Rubrics: Compiling Rubrics into Typed Evaluation Graphs for LLM Judges

arXiv:2608.12097v1 Announce Type: new Abstract: Rubric-based evaluators commonly treat rubrics as prompt context or flat criteria: they specify what to judge but leave criterion composition implicit,

Measure, Don't Optimize: Forecasting Recovery in LLM Unlearning

arXiv:2608.11408v1 Announce Type: new Abstract: Prior white-box studies show that large language models can retain latent traces of target knowledge after unlearning, even when the knowledge is no lon

researcharxiv-cs-cl

The Wording Effect: Quantifying Two-Way Drift in LLM Benchmark Performance

arXiv:2608.11694v1 Announce Type: cross Abstract: A benchmark score comes from a single phrasing of each problem. That single phrasing is treated as if it stood for the whole space of ways the same pr

Who Thinks Best Depends on How Long You Let Them: Budget-Dependent Rankings in LLM Evaluation

arXiv:2608.12150v1 Announce Type: new Abstract: Standard evaluation of large language models assumes stable model rankings across inference conditions. We challenge this assumption by varying the toke

ImpactHO: Importance-Aware KV Cache Transfer for Multi-User Edge LLM Handover

arXiv:2608.10545v1 Announce Type: cross Abstract: Edge LLMs must preserve inference continuity when a user hands over between edge nodes, requiring key-value (KV) cache transfer to the target node. Ho

Is This Your Final Answer? Cross-Contextual Consistency as a Measure of LLM Credibility

arXiv:2608.10315v1 Announce Type: cross Abstract: Large language models (LLMs) are powerful black-box systems, making it difficult to discern whether their answers reflect stable internal beliefs or s

Rethinking LLM Verification: Evidence Structure, Uncertainty, and Selective Refinement

model-releasesarxiv-cs-cv

arXiv:2608.10725v1 Announce Type: new Abstract: Large language models (LLMs) often rely on shortcuts rather than systematic reasoning, raising safety concerns in medical applications. Allowing models

Towards Efficient Reasoning in LLM-Based Recommender Systems via Model Merging

arXiv:2608.10447v1 Announce Type: cross Abstract: Large language model-based recommender systems are increasingly adopting slow-thinking models that generate step-by-step reasoning before making predi

V-FiLLM: Verified Financial LLM Reasoning Benchmark

arXiv:2608.11047v1 Announce Type: new Abstract: While existing benchmarks have made substantial progress in evaluating LLMs across STEM domains, financial reasoning over structured data remains compar

An Expectation-Maximization Perspective on Reinforcement Learning for LLM Reasoning

arXiv:2504.18587v2 Announce Type: replace-cross Abstract: Reinforcement learning has emerged as a powerful approach for improving the reasoning capabilities of large language models, as demonstrated b

Beyond Solvability: Task Learnability as a Static Prior for LLM RL Post-Training

arXiv:2608.09217v1 Announce Type: cross Abstract: Reinforcement learning (RL) has become a central post-training paradigm for eliciting reasoning capabilities in large language models, yet uniform tas

Can LLM Agents Stick to the Script? A Benchmark for Long-Horizon Consistency in Interactive Narratives

arXiv:2608.08160v1 Announce Type: cross Abstract: The rapid advancement of Large Language Models (LLMs) is revolutionizing AI for Games by enabling open-ended and fluid interactive storytelling. Howev

Capability Is Not Propensity: Measuring Pressure-Robust Cooperative Behavior in Civic LLM Agents

arXiv:2608.09485v1 Announce Type: new Abstract: Cooperative capabilities in language models are dual-use. The same social reasoning that supports civic deliberation can also enable strategic omission,

Agents

Controlled Memory Interference in Continual LLM Agents

arXiv:2608.07622v1 Announce Type: new Abstract: Long-term memory enables AI agents to maintain continuity across sessions, personalize behavior, and evolve through accumulated experience. Yet memory e

agentsarxiv-cs-ai

Decoding-Level Taboo: A Diagnostic Stress Test for LLM Robustness

model-releasesarxiv-cs-cl

arXiv:2608.09900v1 Announce Type: new Abstract: Large language model evaluations typically focus on performance under nominal conditions, creating an illusion of capability where models comfortably wa

Failure-Aware Long-Form Translation: Design and Implementation of a Recoverable LLM Translation System

arXiv:2608.09187v1 Announce Type: new Abstract: A long-form translation request can succeed at the API layer and still produce an unusable result. The output may be empty, truncated, filtered, dominat

researcharxiv-cs-cl

From Evaluated Models to Evaluation Aids: A Multi-Evidence Study of LLM-Based Difficulty Calibration for Programming Examinations

arXiv:2608.07523v1 Announce Type: cross Abstract: Difficulty differences across parallel-class programming examinations affect the fairness of course assessment. This study repositions large language

Gaming Without an Attacker: Benchmark Fingerprinting in LLM-Driven Search Under Selection Pressure

arXiv:2608.08722v1 Announce Type: cross Abstract: Benchmarks for systems that are optimized against the evaluation signal measure something different from what they claim. We document this concretely

Applications

H2: A Dual Hybrid Semantic Data Lake Architecture for Medical Data Harmonization with Human-In-the-Loop verified, LLM Driven Metadata Annotation System

arXiv:2608.08056v1 Announce Type: new Abstract: Medical data, by its nature, exhibit a high degree of heterogeneity on multiple levels ranging from (a) different modalities like images, text and time

applicationsarxiv-cs-ai

How Much Backtracking is Enough? Exploring the Interplay of SFT and RL in Enhancing LLM Reasoning

arXiv:2505.24273v2 Announce Type: replace Abstract: Recent advancements in large language models (LLMs) suggest that reinforcement learning (RL) effectively internalizes search strategies, yielding si

Instability of LLM Pre-Pretraining: It Doesn't Always Help. An Investigation on Multiple Languages

model-releasesarxiv-cs-cl

arXiv:2608.08800v1 Announce Type: new Abstract: Pretraining LLMs on artificial languages ('pre-pretraining') is a technique that could reportedly increase token efficiency by 33%, i.e., save up to 33%