1
阿里千问办公上线 Qwen3.8-Flash 及标准模式：单任务生成速度提升约 100%，Token 消耗平均减少 75%
IT之家 8 月 27 日消息，今天，阿里旗下千问办公首发上线刚刚发布的 Qwen3.8-Flash 模型，同时推出标准模式。 即日起，所有用户可通过全新的标准模式体验 Qwen3.8-Flash 。 基于最新的模型，用户可以用更少的积分消耗、更快的 Token 吞吐速度完成任务。未来，千问办公的模…
2
Labeled-Data-Free Meta-Learning: Efficient Task Generation Using Pre-trained Models and Unlabeled Data
不用标注数据就能生成训练任务，这套元学习新方法让模型学会举一反三更省力。
arXiv:2607.02850v1 Announce Type: new Abstract: Meta-learning without labeled data is crucial for real-world applications, where obtaining labeled dat…
3
Confidence-Orchestrated Self-Evolution against Uncertain LLM Feedback
应对不确定LLM反馈的自信编排自我进化方法，减少人工监督依赖，提升模型推理能力
arXiv:2605.28010v1 Announce Type: new Abstract: Self-evolving large language models (LLMs) learn by generating their own training tasks and solutions,…
4
PGT: Procedurally Generated Tasks for improving visual grounding in MLLMs
用程序化生成任务打破MLLM细粒度理解瓶颈，PGT框架一举两得。
arXiv:2605.23883v1 Announce Type: cross Abstract: Despite remarkable progress in Multimodal Large Language Models (MLLMs), these models still struggle…
5
PopuLoRA: Co-Evolving LLM Populations for Reasoning Self- Play
大模型自对弈新范式：群体进化生成可解任务，驱动推理能力持续攀升
Article URL: https://vmax.ai/team/populora-co-evolving-llm-populations-for-reasoning-self-play Comments URL: https://news.ycombinator.com/item?id=4821…