搜索与发现
查事件、来源、文档和知识
统一搜索会同时返回事件、原始来源、采集文档和知识节点。结果按匹配度、来源强度、风险和确认状态排序。
IEEE Transactions on Affective Computing
research_literature · reliability A
匹配来源名称、类型、URL 或可靠性等级。Max Out GRPO Signal: Adaptive Trace Prefix Control for Hard Reasoning Problems
Group Relative Policy Optimization (GRPO) stalls on a model's hardest problems: when no rollout in a group succeeds, the group-relative advantages vanish and the problem contributes no gradient, wasting the frontier examples we most want to learn from. Prepending a correct prefix of a reference solution raises the success rate, making prefix length a continuous knob on difficulty. Concurrent methods set the knob once; AdaPrefix-GRPO turns it into a feedback controller: throughout training it adjusts how much of the solution each problem gets, holding its success rate near 50%, where GRPO's gra
匹配采集文档中文标题、摘要、原始来源或文档类型。An expanding human footprint drives escalating human-elephant conflict across a transboundary African landscape through 2085.
PNAS Nexus; PubMed ID 13339083.
匹配采集文档中文标题、摘要、原始来源或文档类型。How Data Shapes RoPE Frequency Usage: From Positional Scale Matching to Length Generalization
Rotary Position Embeddings (RoPE) provide transformers with a fixed grid of positional frequencies, yet trained models use these frequencies highly non-uniformly. We study what determines this frequency usage and propose a data-centered explanation: RoPE frequencies are selected to match the relative-distance structure of the training data. Viewing each frequency as a positional lens, we formalize a field-resolution tradeoff and show that, for a data-induced dependency profile of width $W$, the optimal frequency scales as $1/W$. This frequency-matching principle explains controlled observation
匹配采集文档中文标题、摘要、原始来源或文档类型。Build a serverless image editing agent with Amazon Bedrock AgentCore harness
This post walks through building a serverless image editor where users upload a photo, describe an edit in plain English, and receive the result in seconds. The agent runs on AgentCore harness without custom orchestration code. We deploy the full solution, including authentication, encrypted storage, three image editing tools, and a React frontend, with a single deployment command. The infrastructure is defined using AWS Cloud Development Kit (AWS CDK).
匹配采集文档中文标题、摘要、原始来源或文档类型。监管路径
把试验阶段、主要终点、安全性、适应症和监管沟通分开看，不把临床更新等同于疗效结论。
匹配知识节点、相关术语或方法论说明。情感计算边界
情绪识别和陪伴产品应区分体验优化、心理健康声明、敏感数据处理和独立证据。
匹配知识节点、相关术语或方法论说明。Co-LMLM: Continuous-Query Limited Memory Language Models
Limited memory language models (LMLMs) externalize factual knowledge during pretraining to a knowledge base (KB), rather than memorizing it in their weights. During generation, the model then fetches knowledge from the KB as needed. This recently introduced paradigm provides multiple advantages, including knowledge control capabilities that remain beyond conventional LLMs. We propose continuous-query LMLM (CO-LMLM), where the KB pairs continuous keys with textual knowledge values, a significant departure from prior reliance on relational KB and queries. CO-LMLM generates flexible vector querie
匹配采集文档中文标题、摘要、原始来源或文档类型。Build an AI-powered AWS support companion with Amazon Bedrock AgentCore
In this post, you build an AWS Support Companion using Amazon Bedrock AgentCore. The agent uses Strands Agents as the orchestration framework and connects to AWS services through the Model Context Protocol (MCP). By the end, you have a working agent that can analyze CloudWatch logs, search AWS documentation, query community knowledge from AWS re:Post, and create support cases, all from a single conversational interface. The solution deploys with a single script using AWS CloudFormation and includes a web frontend built on AWS Amplify for interacting with the agent.
匹配采集文档中文标题、摘要、原始来源或文档类型。肿瘤治疗方案的急性风险监测
临床试验：NCT07601802｜癌症，急性护理服务使用
匹配采集文档中文标题、摘要、原始来源或文档类型。STAT+：美国法官阻止科罗拉多州对 Amgen 药物实施首创价格上限
一名联邦法官阻止科罗拉多州小组对 Amgen 销售的重磅药物 Enbrel 设定价格上限。
匹配采集文档中文标题、摘要、原始来源或文档类型。陪伴依恋风险
关注用户依恋、未成年人、退出机制、数据留存和人机关系边界。
匹配知识节点、相关术语或方法论说明。Does Bielik Know What It Doesn't Know? Activation Dispersion Separates Entity Familiarity from Factual Reliability Across Model Scale
Large language models hallucinate most about entities they have never seen. We ask whether a model's activations betray entity familiarity before a single answer token is generated, and whether that signal predicts the factual reliability of the answers. On four Polish Bielik models (1.5B-11B parameters), we probe four entity domains (athletes, cities, writers, musicians), each with 42 well-known, 42 obscure-but-real, and 42 fabricated entities addressed by a one-sentence question (504 prompts per model). Two unsupervised, single-forward-pass dispersion measures over post-SwiGLU MLP activation
匹配采集文档中文标题、摘要、原始来源或文档类型。A hierarchical memory architecture overcomes context limits in long-horizon multi-agent computational modeling
Large language models (LLMs) demonstrate remarkable reasoning capabilities, yet their stateless architecture fundamentally limits deployment in long-horizon research workflows requiring multi-session continuity and quantitative rigor. Here we present Ensemble QSP, a multi-agent framework featuring a three-layer hierarchical memory architecture that keeps injected context bounded and constant in project duration (mid-term project state: median 301 tokens, max 4,050, across 104 runs) by capping each state category and evicting completed work, enabling continuous autonomous operation without cont
匹配采集文档中文标题、摘要、原始来源或文档类型。PeTeR: Post-Training Robustification of Probabilistic Circuits
Probabilistic circuits (PCs) can model complex joint distributions while supporting exact and efficient computation of many inference queries. However, standard likelihood-based PC learning is vulnerable to overfitting and fragile generalization when confronted with data noise, small sample sizes, or distribution shifts. This can be mitigated using distributionally-robust optimization which consider worst-case distributions within a Wasserstein ball of the empirical distribution, but current methods are limited to training a model from scratch in this framework. Instead, we propose PeTeR: a no
匹配采集文档中文标题、摘要、原始来源或文档类型。Institutional Red-Teaming: Deployment Rules, Not Just Models, Causally Shape Multi-Agent AI Safety
We introduce institutional red-teaming, an evaluation methodology for testing deployment rules in multi-agent AI: hold the agents, objectives, and task state fixed, vary only one rule, and attribute the resulting change in collective behavior to that rule. We instantiate the methodology in IABench-CA, a consequence-allocation benchmark spanning 228 contexts, five canonical rules, and seven model populations (33,924 games), with a normative cooperative reference and auto-labelled reasoning traces. Three findings emerge. (1) Deployment rules causally alter collective safety: changing only the co
匹配采集文档中文标题、摘要、原始来源或文档类型。Laurie Santos: Americans and the Pursuit of Happiness
On “The Interview,” Laurie Santos, a cognitive scientist and a professor at Yale, says that Americans think about happiness in unique ways, and they have for a long time.
匹配采集文档中文标题、摘要、原始来源或文档类型。Flint: A visualization language for the AI era
Short chart specifications are easy to write, but often produce uninspiring results. Flint is an open-source visualization language that offers a middle path, letting AI agents create expressive charts from compact, human-editable specifications. The post Flint: A visualization language for the AI era appeared first on Microsoft Research .
匹配采集文档中文标题、摘要、原始来源或文档类型。Agon: Competitive Cross-Model RL with Implicit Rival Grading of Reasoning
Reinforcement learning from verifiable rewards (e.g. GRPO) is the engine behind today's reasoning models, yet it grades only the final answer. On hard problems this trains models to write more rather than to think better, since the trace itself is never graded and no label for good thinking exists. We introduce Agon, which makes two competing models each other's graders. Both attempt the same problem; in alternating roles, one drafts a solution and the other reads it while solving, and each is rewarded for out-solving the other. To win, a model must out-reason a rival that has seen its work, s
匹配采集文档中文标题、摘要、原始来源或文档类型。STAT+: Pharmalittle: We’re reading about a Vera kidney drug approval, a U.K. Enhertu pricing deal, and more
The FDA approved a kidney disease drug from Vera Therapeutics called Trutakna, to treat IgA nephropathy
匹配采集文档中文标题、摘要、原始来源或文档类型。