探索全球前沿学术脉络

CheckMIABench: Firm Foundations For Membership Inference Attacks on Language Models

arXiv:2606.17464v1 Announce Type: new Abstract: Membership inference attacks (MIAs) are a canonical way to assess a machine learning model's privacy properties. Although several attempts have been made to evaluate MIAs on language models, the extant literature has suffered numerous difficulties in constructing clean evaluations to test new techniques. In particular, subtle distribution shifts between member and non-member sets can undermine the statistical validity of MIAs; recent work has underscored this by showing that "blind" methods with no access to the underlying model can perform far better than published methods on the same benchmarks. This paper constructs a benchmark for principled evaluation of MIAs against LLMs, by leveraging the insight that training data before and after a fixed point during training are drawn from the same distribution. Therefore, all open-source models with intermediate checkpoints and public training data can be converted into MIA testbeds. We apply our framework to a half-dozen published attacks on the Pythia and OLMo family of models, from 70M to 7B parameters. To facilitate further privacy research, we open-source a modular library for designing and implementing attacks in this setting: https://github.com/safr-ai-lab/pandora_llm.

Unbiased Derivative Estimation for Stationary Mean of Parameterized Markov chains

arXiv:2606.11487v1 Announce Type: cross Abstract: We propose a new approach to unbiased estimation of the gradients of the stationary means associated with parametrized families of Markov chains. Our estimators are particularly efficient when the Markov chains have slow mixing rate. Our approach does not require a specific parametrization except for an oracle to evaluate the transition density and its gradient at a given data point without any additional knowledge about the density function itself. It makes our estimator suitable for parametrizations associated with neural networks. The estimator can potentially achieve large improvement in terms of efficiency. Numerical experiments confirm the good performance predicted by the theory.

Nemotron 3 Ultra: Open, Efficient Mixture-of-Experts Hybrid Mamba-Transformer Model for Agentic Reasoning

We introduce Nemotron 3 Ultra, a 550 billion total and 55 billion active parameter Mixture-of-Experts Hybrid Mamba-Attention language model. We pre-trained Nemotron 3 Ultra on 20 trillion text tokens, then extended the context length to 1M tokens, and post-trained using Supervised Fine Tuning (SFT), Reinforcement Learning (RL), and Multi-teacher On-Policy Distillation (MOPD). Nemotron 3 Ultra is our most capable model yet, employing multiple key technologies - LatentMoE, Multi Token Prediction (MTP), NVFP4 pre-training, multi-environment RLVR, MOPD, and reasoning budget control. Nemotron 3 Ultra achieves up to ~6x higher inference throughput as compared to state-of-the-art publicly available LLMs while attaining on-par accuracy. The state-of-the-art accuracy, high inference throughput, and 1M token context length make Nemotron 3 Ultra ideal for long-running autonomous agentic tasks. We open-source the base, post-trained, and quantized checkpoints, along with the training data and recipe on HuggingFace.

CoRe: A Continuously Reward-Finetuned LLM Query Rewriter for Multi-Stage Context-Aware Relevance in Web-Scale Video Search

LLM-based query rewriters in production face a tension: the training reward must reflect how the rewrite is consumed by the production ranker, yet the training procedure must be cheap enough to support continuous redeployment as data drifts. We present CoRe (Context Relevance), such a system, redeployed weekly for over five months in a major short-video search engine. Our reward uses the deployed multimodal relevance model as its source and a multiplicative ratio form mirroring the production fusion algebra, closing the simulation-production gap that offline reward proxies leave open. A semi-online Mixed Preference Optimization loop makes this reward affordable at multi-million-instance weekly scale: a DPO-style pairwise objective restricts the gradient pass to a small top-k/bottom-k subset of sampled trajectories, and a phase structure reduces trainer/inference-server parameter syncs from per-step to per-phase. An automated promotion gate over reward-like and stability metrics detected and recovered from a real reward-hacking incident in production. Rewriter output is consumed as parallel relevance signals at recall, rawrank, and finerank without displacing the original signals, bounding rewriter-failure blast radius. Online A/B from two sequential production launches, first deploying the rewriter at finerank, then extending consumption to recall and rawrank, delivers statistically significant reductions in change-query rate on rewrite-impacted queries, with all headline relevance and engagement metrics moving in the expected direction.

Agents' Last Exam

Recent AI systems have achieved strong results on a wide range of benchmarks, yet these gains have not translated into economically meaningful deployment across many professional domains. We argue that this gap is largely an evaluation problem: widely used benchmarks lack sustained performance measurement on real and economically valuable workflows. This paper introduces Agents' Last Exam (ALE), a benchmark designed to evaluate AI agents on long horizon, economically valuable, real world tasks with verifiable outcomes. Developed in collaboration with 250+ industry experts, ALE covers non-physical industries defined with reference to O*NET / SOC 2018 (the U.S. federal occupational taxonomy). It is organized around a task taxonomy with 55 sub fields grouped into 13 industry clusters covering 1K+ tasks. Current results show that the hardest tier remains far from saturated: across mainstream harness and backbone configurations, the average full pass rate is below 1%. ALE is designed as a living benchmark: its task pool grows continuously as new workflows and industries are onboarded. More broadly, ALE is intended not merely as another leaderboard, but as an instrument for closing the gap between benchmark success and GDP relevant impact.