👍 554
08/09 20:00
We introduce BDH-CQ, a reasoning model that combines in-context learning with recurrent latent reasoning. Inputs presented at inference time continuously update the model's recurrent memory; the model then solves a query through iterative computation in a high-dimensional latent space, without verba
中文介绍 BDH-CQ 是一种结合情境学习与递归潜在推理的推理模型,通过在推断时不断更新模型的递归记忆,以高维潜在空间中的迭代计算解决查询问题。该方法在不依赖显式文本的情况下进行推理推演,为提升推理模型的灵活性和适应性奠定了基础。意义:为推广递归推理方法在智能体与推理任务中的应用提供新的思路。
👍 185
08/08 20:00
On-policy (Self-)Distillation (OPD / OPSD) has shown strong potential for post-training large language models (LLMs). However, existing methods still rely heavily on external supervision, including ground-truth signals, environmental feedback, or guidance from larger models, and therefore fall short
中文介绍 本文提出了一种无监督的在职自蒸馏方法(OPSD),旨在提升大型语言模型(LLMs)的后训练效果。该方法不依赖于外部监督信号,而是专注于内部自我优化的潜力,通过内在信号进行模型精炼,为无监督学习提供了一种新型的解决方案。意义:推动无监督学习在语言模型领域的研究,扩大其在各类应用中的应用潜力。
👍 174
08/10 20:00
After an older adult misses a medication dose, a software agent can send another reminder and an embodied agent can bring the medication. Yet neither explains whether the person forgot, is confused, has side effects, or deliberately refused, nor what support is appropriate. This reveals a structural
中文介绍 ComBodied Agents 提出了一个以人为中心的智能体新范式,旨在加强对老年人药物服用情况的理解和支持。该方法通过分析用户行为和背景信息,适应性地提供个性化支持,提高医疗提醒的有效性,为智能体在健康管理领域的应用提供新的视角。意义:改善智能体在人机交互与健康管理中的应用,推动人机协作的深入发展。
👍 116
08/09 20:00
Agentic systems are increasingly expected to improve after deployment, yet single-entity self-evolution is often bounded by a static learning context, such as fixed tasks and feedback. This survey focuses on co-evolution in agentic systems, a multi-component form of self-evolution in which multiple
中文介绍 本文调查了智能系统的共同演化,提出了多组分的自我演化框架,旨在打破传统单一实体学习的限制。通过分析各组件间的相互影响,该研究为智能系统的动态适应与协同演进提供了理论基础和实践启示。意义:促进智能系统在复杂环境下的自适应能力,为未来智能体算法的开发指明方向。
👍 86
08/09 20:00
Leading large language model providers now conceal their models' step-by-step reasoning, or chain-of-thought, to protect intellectual property and limit information leakage. Rather than storing these traces server-side, providers return them to the client as blocks of encrypted text, which the clien
中文介绍 研究揭示了如何从专有语言模型的 API 中提取推理轨迹,面对知识产权保护带来的挑战,通过对加密文本的分析获取模型推理过程,从而生成更可靠的推理结果。该方法为解析大型语言模型的内在机制提供了可能的路径,具有重要的实用价值。意义:为理解和改进推理模型的透明度和可解释性铺平了道路。
👍 47
08/06 20:00
Memory systems have shown promise for improving agent performance, but their potential remains largely unexplored for small language models, which struggle to generate sufficient successful trajectories on their own. We propose Agent Memory Distillation (AMD), a training-free framework that transfer
中文介绍 Agent Memory Distillation(AMD)是一种新提出的框架,用于提升小型语言模型在生成成功轨迹方面的能力。该方法不需额外的训练,通过传递层次性记忆来增强智能体的表现,展现了记忆系统在小型模型中的潜力,帮助其实现更复杂的任务执行。意义:推动小型语言模型的性能提升,增强其在复杂应用中的适应性。
👍 27
08/02 20:00
Conversational assistants increasingly recommend follow-up edits to help users continue a task. Existing systems primarily target text-only interactions, leaving image-creation conversations underexplored. In image-creation tasks, useful follow-up edit suggestions must reflect user preferences, offe
中文介绍 研究提出了一种针对图像编辑任务的对话系统,旨在根据用户偏好推荐后续编辑操作。通过结合用户输入和任务背景,该系统能够提供更符合用户需求的建议,从而改善用户的编辑体验。意义:推动对话系统在多模态互动中的应用,为图像生成与协作编辑提供支持。
👍 24
08/07 20:00
Large language model (LLM) inference serving is increasingly constrained by memory rather than compute. As long-context and long-form reasoning workloads become more prevalent, the key-value (KV) cache dominates both memory footprint and memory traffic during LLM token generation, i.e., decode. In p
中文介绍 OasisKV 提出了一种新的关键值缓存策略,旨在提升大型语言模型推理的性能。通过前瞻性稀疏预取,显著减少了内存占用与访问时间,以适应复杂长上下文的生成需求。该方法为高性能推理提供了可行的解决方案。意义:为解决大规模模型推理中内存瓶颈提供了创新思路,推动高效AI应用的发展。
👍 22
08/06 20:00
Self-improving coding agents that iteratively rewrite their own source code have demonstrated impressive performance on coding tasks. However, existing solutions generally derive self-modification from a single failure trajectory at a time, overlooking rich comparative signals available in the agent
中文介绍 Mendel Gödel Machine 介绍了一种基于比较演化的递归自我改进编码智能体的方法,该方法通过多个失败实例的对比,利用丰富的信号实现智能体的自主优化,显著提升了其在编码任务中的表现。意义:为自我改进智能体的发展提供了新颖的理论基础和实践指导,推动智能体在动态环境中的适应性提升。
👍 16
08/08 20:00
Running a business is a challenging form of intelligent work. Operators must infer opportunities from partial signals, commit capital under uncertainty, adapt to delayed outcomes in a changing market, and satisfy regulatory obligations before trading legally. Frontier LLM agents can increasingly com
中文介绍 Business Arena 研究了大型语言模型(LLM)在逼真市场环境中的业务操作能力,拓展了传统LLM的评估与应用。这项研究关注在复杂市场中运用智能体进行决策的能力,与不确定性和延迟结果的应对策略。意义:为智能体在商业决策和复杂环境中的应用奠定了基础,推动未来经济智能系统的发展。
👍 15
08/10 20:00
Large language model (LLM) agents are increasingly deployed as personal assistants. Existing evaluations, however, mostly use short, self-contained requests in static environments. Everyday life assistance is different. A task runs for weeks rather than minutes. The world keeps changing while the ag
中文介绍 VibeLifeBench 评估了大型语言模型(LLM)在动态日常生活中的表现,特别是与用户在长期任务中持续交互的能力。与静态环境下的短期任务评估不同,该研究关注智能体如何在不断变化的环境中提供持续支持。意义:促进AI智能体在日常生活辅助中的实用性,推动个人助手领域的发展。
👍 12
08/09 20:00
Large language model evaluations typically focus on performance under nominal conditions, creating an illusion of capability where models comfortably walk a narrow, highly optimized generation corridor. In real-world deployments, however, complex system prompts, safety guardrails, and structural con
中文介绍 本文提出了一种针对大型语言模型的解码级别压力测试(Taboo),旨在评估模型在复杂提示及安全防护下的表现。通过模拟真实场景,检测模型在常规条件下的潜在脆弱性,为AI系统的稳健性提升提供了重要参考。意义:增强大型语言模型在现实应用中的可靠性与实用性,推动对抗性安全研究的发展。
👍 11
08/08 20:00
Long-horizon research agents solve open-ended tasks through iterative retrieval, aggregation, and synthesis, but context grows rapidly while the marginal value of additional evidence often declines. This leads to unnecessary token cost, higher latency, and noisier inputs for final report generation.
中文介绍 研究探讨了高层次研究智能体如何通过迭代检索、聚合和综合解决开放性任务。面对上下文增长和附加证据的边际价值下降,该方法提出减少不必要的计算与时间成本的方法,以提高效率。意义:推动研究智能体在长期任务中的优化,提升其在复杂情境下的应用表现。
👍 11
08/10 20:00
Self-evolving agents accumulate reusable skills by appending successful procedures and failure fixes. Over time, the same requirement is often restated in several branches, examples, and warnings, while common action sequences are copied rather than reused. The resulting skill becomes expensive to i
中文介绍 SkillZip 提出了无评估技能压缩的方案,帮助自我进化的智能体更有效地积累可复用技能。研究揭示了智能体在多个领域重复需求下的技能利用不足,为未来智能体技能管理提供了有价值的思路。意义:为智能体技能复用与管理提供新的理解框架,推动智能体技术的演进。
👍 11
08/09 20:00
General-purpose reward models are increasingly the bottleneck for scaling robot learning, yet the recipe for learning value-related capabilities from large-scale heterogeneous corpora remains underexplored. Existing approaches tie supervision to task-internal anchors such as preferences or normalize
中文介绍 RynnValue 研究了通用奖励模型在机器人学习中的瓶颈,提出了一种通过时间距离来提升价值学习能力的方法。该方法能够从异构数据集中学习价值相关能力,为机器人提供更为高效的学习机制。意义:推动机器人技术领域的进步,加强自主学习与适应能力的研究。
👍 10
08/10 20:00
Long-document understanding often requires reasoning over many visually rich pages, making inference costly and prone to context rot. In this work, we propose InSight-doc, an agentic visual perception framework that treats visual resolution as an adaptive reasoning-time resource. InSight-doc starts
中文介绍 InSight-doc 提出了一种针对长文档理解的智能视觉感知框架,旨在优化推理过程中的视觉资源利用。该框架将视觉分辨率视为动态推理时间资源,有助于提升长文本处理的效率和准确性。意义:为文档理解与推理的融合研究提供新方向,推动智能体在复杂文档处理中的应用发展。
👍 10
08/10 20:00
Visual document retrieval (VDR) is dominated by multi-billion-parameter models that are slow to index at full corpus scale and expensive to serve. Prior compression routes either train a smaller multi-vector encoder from scratch or distil only the query side; neither yields a compact single-vector r
中文介绍 DistilVDR 提出了一种紧凑型视觉文档检索系统,通过双学生蒸馏方法实现更高效的文档检索。相较于传统多参数模型,该方案显著降低了检索成本并提高了系统响应速度,为视觉检索技术的轻量级应用提供了解决方案。意义:推动视觉文档搜索技术的高效化,助力多模态信息检索的发展。
👍 10
08/10 20:00
We study reference-free post-training for multilingual machine translation with open large language models. Starting from the supervised-finetuned MiLMMT-46-v0.1 models, we apply Group Relative Policy Optimization (GRPO) with a reward that averages two reference-free quality estimation models and is
中文介绍 本文探讨了针对多语言机器翻译的无参考后训练方法,通过对大型语言模型进行优化,不依赖实际参考进行质量评估,提升多语言模型的翻译效果。该方法标志着机器翻译领域新的优化方向,推动多语言处理的技术进步。意义:加深多语言模型在真实应用场景中性能提升的研究,推动机器翻译系统的实用化。
👍 10
08/10 20:00
Large language models (LLMs) are increasingly deployed as mobile assistants, where a key challenge is leveraging personal information scattered across multiple applications (apps) to complete user instructions. However, due to the lack of dedicated benchmarks, their capabilities remain poorly unders
中文介绍 SPIEval 针对大型语言模型(LLM)作为移动助手的能力进行评估,重点分析如何利用分散在多个应用中的个人信息以完成用户指令。该研究通过建立专用基准,揭示了当前模型在处理分散信息时的局限性,为提高其综合应用能力提供了参考。意义:提升AI助手的个人化服务能力,加强用户与模型之间的交互效果。
👍 10
08/09 20:00
With the rapid advancement of large language models (LLMs), harnesses have become essential infrastructure for deploying agents across a wide range of domains. The fast-evolving harness ecosystem has also made rigorous capability evaluation increasingly important. However, efficiently building an en
中文介绍 A^2E 是一种端到端的代理审计引擎,旨在支持大型语言模型的高效部署与能力评估。通过提升审计手段的严谨性,该工具有效优化了智能体在各种领域的部署效果,推动了智能体开发的规范化进程。意义:为智能体的系统性评估与优化提供支持,推动AI技术的全面发展。