Search: “llm”

Curriculum Learning for LLM Pretraining: An Analysis of Learning Dynamics

arXiv:2601.21698v2 Announce Type: replace-cross Abstract: Curriculum learning changes the order of pretraining data, but it remains unclear how ordering changes the learning dynamics. We pretrain mode

Content type

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

DiagnosticIQ: A Benchmark for LLM-Based Industrial Maintenance Action Recommendation from Symbolic Rules

DGX agent

arXiv:2605.08614v1 Announce Type: new Abstract: Monitoring complex industrial assets relies on engineer-authored symbolic rules that trigger based on sensor conditions and prompt technicians to perfor

model-releasesarxiv-cs-ai

12 May 2026

Different Prompts, Different Ranks: Prompt-aware Dynamic Rank Selection for SVD-based LLM Compression

arXiv:2605.08568v1 Announce Type: new Abstract: Large language models (LLMs) have rapidly grown in scale, creating substantial memory and computational costs that hinder efficient deployment. Singular

Tutorials

Dual Tuning for Reasoning Efficacy-Driven Data Curation in Multimodal LLM Training

arXiv:2603.04415v2 Announce Type: replace Abstract: Reasoning post-training improves Large Language Models (LLMs) on complex tasks such as mathematics and coding, but its benefits across diverse multi

tutorialsarxiv-cs-cl

Efficient Evaluation of LLM Performance with Statistical Guarantees

arXiv:2601.20251v3 Announce Type: replace-cross Abstract: Exhaustively evaluating many large language models (LLMs) on a large suite of benchmarks is expensive. We cast benchmarking as finite-populati

Efficient LLM Collaboration via Planning

arXiv:2506.11578v4 Announce Type: replace Abstract: Recently, large language models (LLMs) have demonstrated strong performance, ranging from simple to complex tasks. However, while large models achie

local-aiarxiv-cs-ai

From Detection to Recovery: Operational Analysis on LLM Pre-training with 504 GPUs

arXiv:2605.09370v1 Announce Type: cross Abstract: Large-scale AI training is now fundamentally a distributed systems problem, and hardware failures have become routine operating conditions rather than

local-aiarxiv-cs-ai

GAMBIT: A Three-Mode Benchmark for Adversarial Robustness in Multi-Agent LLM Collectives

model-releasesarxiv-cs-cl

arXiv:2605.09027v1 Announce Type: new Abstract: In multi-agent systems (MAS), a single deceptive agent can nullify all gains of an agentic AI collective and evade deployed defenses. However, existing

Agents

GameChat: Multi-LLM Dialogue for Safe, Agile, and Socially Optimal Multi-Agent Navigation in Constrained Environments

arXiv:2503.12333v2 Announce Type: replace Abstract: Safe, agile, and socially compliant multi-robot navigation in cluttered and constrained environments remains a critical challenge. This is especiall

agentsarxiv-cs-ro

HyPER: Bridging Exploration and Exploitation for Scalable LLM Reasoning with Hypothesis Path Expansion and Reduction

arXiv:2602.06527v2 Announce Type: replace Abstract: Scaling test-time compute with multi-path chain-of-thought improves reasoning accuracy, but its effectiveness depends critically on the exploration-

LLM Translation of Compiler Intermediate Representation

arXiv:2605.08247v1 Announce Type: cross Abstract: GCC and LLVM underpin much of modern software infrastructure, relying on distinct Intermediate Representations (IRs) to drive optimizations and code g

Navigating LLM Valley: From AdamW to Memory-Efficient and Matrix-Based Optimizers

arXiv:2605.09176v1 Announce Type: cross Abstract: Training large language models requires optimization algorithms that are not only statistically effective, but also computationally and memory efficie

Navigating the Sea of LLM Evaluation: Investigating Bias in Toxicity Benchmarks

arXiv:2605.10639v1 Announce Type: new Abstract: The rapid adoption of LLMs in both research and industry highlights the challenges of deploying them safely and reveals a gap in the systematic evaluati

Hardware

Not All Thoughts Need HBM: Semantics-Aware Memory Hierarchy for LLM Reasoning

arXiv:2605.09490v1 Announce Type: new Abstract: Reasoning LLMs produce thousands of chain-of-thought tokens whose KV cache must reside in scarce GPU HBM. The dominant response -- permanently evicting

hardwarearxiv-cs-cl

OPT-BENCH: Evaluating the Iterative Self-Optimization of LLM Agents in Large-Scale Search Spaces

arXiv:2605.08904v1 Announce Type: new Abstract: Large Language Models (LLMs) have demonstrated remarkable capabilities in reasoning and tool use. However, the fundamental cognitive faculties essential

OTora: A Unified Red Teaming Framework for Reasoning-Level Denial-of-Service in LLM Agents

arXiv:2605.08876v1 Announce Type: new Abstract: Large Language Models (LLMs) are increasingly deployed as autonomous agents that execute tool-augmented, multi-step tasks, where latency is a critical f

Rethinking Expert Trajectory Utilization in LLM Post-training for Mathematical Reasoning

arXiv:2512.11470v2 Announce Type: replace-cross Abstract: Supervised Fine-Tuning (SFT) and Reinforcement Learning (RL) dominate the post-training landscape for mathematical reasoning, yet differ funda

researcharxiv-cs-cl

Rethinking RL for LLM Reasoning: It's Sparse Policy Selection, Not Capability Learning

arXiv:2605.06241v2 Announce Type: replace Abstract: Reinforcement learning has become the standard for improving reasoning in large language models, yet evidence increasingly suggests that RL does not

safetyarxiv-cs-cl

Seed Hijacking of LLM Sampling and Quantum Random Number Defense

arXiv:2605.08313v1 Announce Type: cross Abstract: Large language models (LLMs) rely on deterministic pseudorandom number generators (PRNGs) for autoregressive sampling, creating a critical supply-chai

Selective Deficits in LLM Mental Self-Modeling in a Behavior-Based Test of Theory of Mind

arXiv:2603.26089v2 Announce Type: replace-cross Abstract: The ability to represent oneself and others as agents with knowledge, intentions, and belief states that guide their behavior - Theory of Mind

SLIM: Sparse Latent Steering for Interpretable and Property-Directed LLM-Based Molecular Editing

arXiv:2605.10831v1 Announce Type: cross Abstract: Large language models possess strong chemical reasoning capabilities, making them effective molecular editors. However, property-relevant information

Spatial Priming Outperforms Semantic Prompting: A Grid-Based Approach to Improving LLM Accuracy on Chart Data Extraction

arXiv:2605.08220v1 Announce Type: new Abstract: The automated extraction of data from scientific charts is a critical task for large-scale literature analysis. While multimodal Large Language Models (

Statistical Scouting Finds Debate-Safe but Not Debate-Useful Cases: A Matched-Ceiling Study of Open-Weight LLM Reasoning Protocols

model-releasesarxiv-cs-cl

arXiv:2605.09618v1 Announce Type: new Abstract: When should a language model answer directly, sample and vote, or engage in multi-agent debate? Recent work shows voting often explains much of the gain

TAD: Temporal-Aware Trajectory Self-Distillation for Fast and Accurate Diffusion LLM

arXiv:2605.09536v1 Announce Type: cross Abstract: Diffusion large language models (dLLMs) offer a promising paradigm for parallel text generation, but in practice they face an accuracy-parallelism tra

The Geometry of Forgetting: Temporal Knowledge Drift as an Independent Axis in LLM Representations

arXiv:2605.09195v1 Announce Type: new Abstract: Large language models confidently produce outdated answers, and no existing method can detect them. We show this is not an engineering failure but a str

The Silent Vote: Improving Zero-Shot LLM Reliability by Aggregating Semantic Neighborhoods

arXiv:2605.09739v1 Announce Type: cross Abstract: Large Language Models are increasingly used as zero-shot classifiers in complex reasoning tasks. However, standard constrained decoding suffers from a

To Redact, or not to Redact? A Local LLM Approach to Deliberative Process Privilege Classification

arXiv:2605.10211v1 Announce Type: cross Abstract: Government transparency laws, like the Freedom of Information (FOIA) acts in the United States and United Kingdom, and the Woo (Open Government Act) i

Tokens-per-Parameter Coverage Is Critical for Robust LLM Scaling Law Extrapolation

arXiv:2605.08541v1 Announce Type: new Abstract: Neural scaling laws approximate a language model's loss as a power-law function of parameter count N and token count D. Following Chinchilla-style compu

Uncensored LLM

Uncensored LLMs are architectures that have been modified or fine-tuned to remove standard safety alignment layers (guardrails) that limit a model's ability to discuss sensitive topics. Ollama offers

local-air-ollama

When Less is More: The LLM Scaling Paradox in Context Compression

arXiv:2602.09789v3 Announce Type: replace Abstract: Scaling up model parameters has long been a prevalent training paradigm driven by the assumption that larger models yield superior generation capabi

researcharxiv-cs-lg

Adaptive Negative Reinforcement for LLM Reasoning:Dynamically Balancing Correction and Diversity in RLVR

arXiv:2605.07137v1 Announce Type: cross Abstract: Reinforcement learning with verifiable rewards (RLVR) has become a highly effective method for improving the reasoning abilities of Large Language Mod

Best LLM on a 32Gb M5 MBA

With 32GB of RAM, 30B Q4 models run comfortably on M5 MacBook Air , and popular choices include Qwen 3.6 27B (requiring 18GB+ VRAM) and Qwen 3.6 9B (lightweight, runs on 8GB) . Ollama's macOS implemen

local-air-ollama

Beyond LoRA vs. Full Fine-Tuning: Gradient-Guided Optimizer Routing for LLM Adaptation

arXiv:2605.07111v1 Announce Type: cross Abstract: Recent literature on fine-tuning Large Language Models highlights a fundamental debate. While Full Fine-Tuning (FFT) provides the representational pla

CyBiasBench: Benchmarking Bias in LLM Agents for Cyber-Attack Scenarios

arXiv:2605.07830v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly deployed as autonomous agents in offensive cybersecurity. In this paper, we reveal an interesting phenom

From Time Series Analysis to Question Answering: A Survey in the LLM Era

arXiv:2506.11512v2 Announce Type: replace-cross Abstract: Recently, Large Language Models (LLMs) have introduced a novel paradigm in Time Series Analysis (TSA), leveraging strong language capabilities

How Value Induction Reshapes LLM Behaviour

arXiv:2605.07925v1 Announce Type: new Abstract: Conversational Large Language Models are post-trained on language that expresses specific behavioural traits, such as curiosity, open-mindedness, and em

safetyarxiv-cs-cl

Tutorials

LKV: End-to-End Learning of Head-wise Budgets and Token Selection for LLM KV Cache Eviction

arXiv:2605.06676v1 Announce Type: cross Abstract: Long-context inference in Large Language Models (LLMs) is bottlenecked by the linear growth of Key-Value (KV) cache memory. Existing KV cache compress

tutorialsarxiv-cs-cl

Agents

LLM-Guided Open Hypothesis Learning from Autonomous Scanning Probe Microscopy Experiments

arXiv:2605.06839v1 Announce Type: cross Abstract: Autonomous experimentation has transformed microscopy and materials discovery by enabling closed-loop optimization including imaging and spectroscopy

agentsarxiv-cs-ai

MatryoshkaLoRA: Learning Accurate Hierarchical Low-Rank Representations for LLM Fine-Tuning

arXiv:2605.07850v1 Announce Type: cross Abstract: With the rise in scale for deep learning models to billions of parameters, the computational cost of fine-tuning remains a significant barrier to depl

Agents

MedExAgent: Training LLM Agents to Ask, Examine, and Diagnose in Noisy Clinical Environments

arXiv:2605.07058v1 Announce Type: cross Abstract: Real-world clinical diagnosis is a complex process in which the doctor is required to obtain information from both interaction with the patient and co

agentsarxiv-cs-ai

MiniAppBench: Evaluating the Shift from Text to Interactive HTML Responses in LLM-Powered Assistants

arXiv:2603.09652v3 Announce Type: replace Abstract: With the rapid advancement of Large Language Models (LLMs) in code generation, human-AI interaction is evolving from static text responses to dynami

MISA: Mixture of Indexer Sparse Attention for Long-Context LLM Inference

arXiv:2605.07363v1 Announce Type: cross Abstract: DeepSeek Sparse Attention (DSA) sets the state of the art for fine-grained inference-time sparse attention by introducing a learned token-wise indexer

POETS: Uncertainty-Aware LLM Optimization via Compute-Efficient Policy Ensembles

arXiv:2605.07775v1 Announce Type: cross Abstract: Balancing exploration and exploitation is a core challenge in sequential decision-making and black-box optimization. We introduce POETS (extbf{Po}licy

Reformulating KV Cache Eviction Problem for Long-Context LLM Inference

arXiv:2605.07234v1 Announce Type: cross Abstract: Large language models (LLMs) support long-context inference but suffer from substantial memory and runtime overhead due to Key-Value (KV) Cache growth

local-aiarxiv-cs-ai

Semantic State Abstraction Interfaces for LLM-Augmented Portfolio Decisions: Multi-Axis News Decomposition and RL Diagnostics

arXiv:2605.06730v1 Announce Type: new Abstract: We introduce Semantic State Abstraction Interfaces (SSAI): a methodological template for mapping sparse unstructured text into K auditable, named coordi

researcharxiv-cs-lg

The Memory Curse: How Expanded Recall Erodes Cooperative Intent in LLM Agents

arXiv:2605.08060v1 Announce Type: cross Abstract: Context window expansion is often treated as a straightforward capability upgrade for LLMs, but we find it systematically fails in multi-agent social

VESPO: Variational Sequence-Level Soft Policy Optimization for Stable Off-Policy LLM Training

arXiv:2602.10693v3 Announce Type: replace-cross Abstract: Off-policy updates are inevitable in reinforcement learning (RL) for large language models (LLMs) due to rollout staleness from asynchronous t

WorldCup Sampling for Multi-bit LLM Watermarking

arXiv:2602.01752v2 Announce Type: replace Abstract: As large language models (LLMs) generate increasingly human-like text, watermarking has emerged as a promising solution for reliable attribution bey

researcharxiv-cs-cl