Tracked topic
Large language models are machine learning models trained to predict and generate text and other language-based outputs.
…Prompt-aware Dynamic Rank Selection for SVD-based LLM Compression (2026) TwinQuant: Learnable Subspace Decomposition for 4-Bit LLM Quantization (2026) Cross-Layer Subspace Coupling for LLM Compression: A Unifying Framework and…
Papers arxiv:2606.00467 On the Limits of LLM Adaptability: Impact of Model-Internalized Priors on Annotation Task Performance Published on May 30 Submitted by Rafal Kocielnik on Jun 12 California institute…
Papers arxiv:2605.28510 Efficient and Scalable Provenance Tracking for LLM-Generated Code Snippets Published on May 27 Submitted by Andrea Gurioli on May 28 Authors: , , , , Abstract A hybrid approach combining vector…
…AI-generated summary Industrial LLM agent systems often separate planning from execution, yet LLM planners frequently produce structurally invalid or unnecessarily long workflows, leading to brittle failures and avoidable tool and API…
Papers arxiv:2606.04978 Probing Outcome-Level Resemblance and Mechanism-Level Alignment in LLM Risk Decisions: Evidence from the St. Petersburg Game Published on Jun 3 Submitted by BruceLyu on Jun 4…
…Existing LLM-based data curation methods primarily rely on human-designed workflows, leaving it unexamined whether LLMs can autonomously execute an end-to-end data engineering pipeline for model specialization . We formalize…
…With the ever-growing list of available LLMs, inference routers are being proposed to select the most appropriate LLM for each prompt. However, existing routing methods either optimize cost across weak-to…
Papers arxiv:2605.30260 How LoRA Remembers? A Parametric Memory Law for LLM Finetuning Published on May 28 Submitted by Ningyu Zhang on May 29 AGI Lab Authors: , , , , , , Ningyu Zhang Abstract Research…
…A Benchmark for Evaluating Multilingual LLM Safety in South Asia (2026) ML-Bench&Guard: Policy-Grounded Multilingual Safety Benchmark and Guardrail for Large Language Models (2026) HarmChip: Evaluating Hardware Security Centric LLM…
…A key idea is LLM-as-Observer: instead of asking an LLM to directly predict the next action, we let a small frozen LLM read the interaction and use its hidden states…
