← Back to all news

Large language models

Model releases, architecture work and prompting research on large language models — from frontier-lab announcements to the arXiv papers behind them.

28,678 stories · RSS feed

arXiv AI
Oct 2

Agent Evaluation Reliability: More Tasks Won't (Always) Fix An Agent Leaderboard

Agent evaluations increasingly benchmark LLMs, but rankings can be swayed by evaluation conditions such as scaffolds or tasks, making reliability claim‑dependent. A Bayesian variance‑decomposition framework applied to 22 benchmarks shows that reliability varies with the measurement goal: fixed model‑scaffold systems rank reliably, while underlying‑model rankings are less stable. Scaffold choice can alter conclusions, and adding more tasks only modestly improves reliability when scaffold coverage is limited; however, pooling diverse benchmarks can substantially raise cross‑task ranking reliability and reduce cost.

By Michael Hardy, Ruhana Azam, Anka Reuel, Mykel Kochenderfer, Sanmi Koyejo
llmsagentsbenchmarks
More like this →
arXiv AI
Oct 2

Backdoor Purification for LoRA-Tuned LLMs via Null-Space Projection

The paper introduces a method to purify LoRA-tuned large language models (LLMs) against backdoor attacks without relying on trigger knowledge, clean references, or retraining. By extracting high‑fidelity backdoor directions and projecting LoRA updates onto orthogonal null spaces in input and output channels, the approach reduces attack success rates from nearly 100% to under 10%. Experiments demonstrate that this null‑space projection preserves both the base model’s general capabilities and the new downstream skills learned through the adapter across various tasks.

By Jianwei Li, Jung-Eun Kim
llmsnlpfine-tuningmultimodalbenchmarks
More like this →
arXiv AI
Oct 2

ReHoPER: Receding-Horizon Planning for Enhanced Reasoning

ReHoPER is an inference‑only, zero‑shot method that enhances large language models’ reasoning by generating and answering intermediate questions along multiple paths before producing a final answer. It plans a horizon of candidate intermediate questions, selects one to answer, and replans based on the updated history. The approach is task‑agnostic, using generic instructions across datasets and models without labeled data or task‑specific prompt design, and it outperforms strong baselines on several datasets, notably achieving the largest gains on the new iLLC benchmark for compositional reasoning.

By Saeed Ahmadnia, Cornelia Caragea
llmsbenchmarks
More like this →
arXiv AI
Oct 2

Distilling Directional Verification

arXiv:2610.00997v1 Announce Type: cross Abstract: Knowledge distillation aims to transfer the factual knowledge of large language models to smaller models for efficient deployment. Yet a teacher may...

By Jungseob Lee, Sugyeong Eo, Seongtae Hong, Seungyoon Lee, Chanjun Park, Jaehyung Seo, Heuiseok Lim
llmsefficiency
More like this →
arXiv AI
Oct 2

Sharpening Tax in Post-Training

The paper investigates how reinforcement learning post‑training of large language models (LLMs) tends to sharpen existing behaviors, improving single‑shot accuracy but reducing solution coverage. It shows that pre‑trained LLMs, when paired with a lightweight inference harness, can outperform post‑trained models in coverage for agentic tasks that require multi‑turn tool use. The authors introduce the Sharpening Tax metric to quantify this trade‑off, analyze its prevalence across 14 model pairs and 42 benchmark cases, and propose a Bayesian sampler, PTGS, that mitigates the tax by adapting sampling temperature to prompt difficulty.

By Changdae Oh, Qi Zeng, Qi Qi, Andrey Zhmoginov, Deren Lei, Yun He, Hoang Phan, Hangoo Kang, Azalia Mirhoseini, Sharon Li
llmsagentsreinforcement-learningbenchmarks
More like this →
arXiv AI
Oct 2

Your Language Model is Its Own Critic: Reinforcement Learning with Value Estimation from Actor's Internal States

The paper introduces POISE, a reinforcement learning algorithm that uses a model’s internal states as a value estimator to reduce variance in reinforcement learning with verifiable rewards (RLVR). By employing a lightweight probe that reads internal signals during the forward pass, POISE predicts baselines online and uses a cross‑rollout construction to keep gradients unbiased. Experiments on Qwen3‑4B and OLMo3‑7B‑Instruct‑DPO across six domains show POISE outperforms existing RLVR baselines, offering more stable training and a value model that generalizes across tasks and scales with the policy.

By Yunho Choi, Jongwon Lim, Woojin Ahn, Minjae Oh, Jeonghoon Shim, Yohan Jo
llmsreinforcement-learning
More like this →
arXiv AI
Oct 2

Reshape and Recur: Improving SSMs with Input Reshaping and Depth Recurrence

The paper proposes two extensions to State Space Models (SSMs) to reduce memory usage and improve performance. First, it introduces depth recurrence, allowing a looped SSM with fewer parameters to match the performance of a larger, non-recurrent model. Second, it advocates using a fixed time granularity across tasks by reshaping input sequences, which enhances how information is presented to the model. Both techniques consistently benefit four representative SSM architectures (LRU, S5, LinOSS, LrcSSM).

By M\'onika Farsang, Ramin Hasani, Daniela Rus, Radu Grosu
llmsragefficiency
More like this →