Pai 的 AI 雷达
2026-08-26：Learning never stops: How AI；Bringing ChatGPT for Teacher；The Hugging Face incident an
LLM agents remain unreliable on long-horizon tasks, where small local failures can compound over extended interactions and lead to overall task failure. Although external harnesses can substantially improve robustness, h
Prompt injection is listed as the \#1 threat to AI agents. When an agent accesses external data from websites, files, or emails, an attacker may inject a prompt into the data, saying, "Ignore all prior instructions and p
Recursive self-improvement (RSI) remains hard in long-horizon tasks, where growing histories obscure the task state and misalign skill invocation. We introduce Recuris, a recursive Experiential-Working Memory architectur
We study autonomous mathematical discovery in the Station, an open-world multi-agent environment in which AI agents from different model families pursue a shared research goal without a central coordinator or scripted pi
Multimodal large language models (MLLMs) have become a prevailing paradigm for unified video perception. However, post-training on large multi-task datasets remains challenging, as existing reinforcement learning methods
Vision-language-action (VLA) models have become a dominant paradigm for generalist embodied agents, demonstrating strong complex and long-horizon task completion in structured settings. Yet it remains an open question wh
Universal multimodal embeddings are becoming a core component of modern AI systems, enabling heterogeneous content to be represented in a shared space for applications such as retrieval, recommendation, classification, a
Reinforcement learning can align diffusion models with human preferences and task-specific objectives, but endpoint rewards do not specify how an intermediate denoising prediction should change. We introduce DiffusionOPS
As large language models (LLMs) continue to advance in coding capabilities, their potential in cybersecurity has drawn increasing research attention, with closed-source LLMs (e.g., Mythos) delivering advanced cybersecuri
Group-based reinforcement learning methods such as GRPO for large language models avoid training a critic by sampling multiple responses for each prompt. A reliable critic could instead estimate token-level advantages fr
Reinforcement Learning with Verifiable Rewards (RLVR) and on-policy distillation (OPD) have become two widely adopted paradigms for post-training large language models. However, RLVR suffers from sparse task-level feedba
We present LAION-BVD, a large-scale open video dataset for multimodal learning, which contains 1.3B platform-specific video URLs collected from CommonCrawl. From these, we download 80M videos with a total duration of 10
Self-improving LLM agents refine answers, not the process that produces those answers. Systems that add a meta-level hold that level fixed, and those that edit themselves must leave part of their own editing machinery un
Video games provide a scalable source of training data for video world models, offering diverse environments, complex interactions, and abundant in-the-wild gameplay videos. However, raw gameplay footage entangles the ga
Smart glasses are evolving from capture and display accessories into first-person intelligence platforms that connect human perception, persistent context, and digital or physical action. Their on-body viewpoint aligns w
Large Language Models excel at code generation, yet competitive programming exposes a persistent failure mode: existing multi-agent pipelines distribute work over generic planner, coder, and debugger roles and delegate t
Concurrent multi-agent coding promises division of labor across modules, robustness through redundancy, and parallel exploration at the natural granularity of multi-file projects. Realtime collaborative editing protocols
Large language model (LLM) agents coordinate complex tasks through multi-role and multi-stage workflows. Upstream state is repeatedly transformed into intermediate language artifacts, such as summaries, plans, tickets, m
Outcome-supervised search agents learn when and how to retrieve evidence, but terminal rewards neither localize intermediate errors nor redirect an ongoing trajectory before those errors compound. Treating corrective fee
LLM-based agents execute multi-step tasks, but their behavioral structure remains opaque: long unstructured traces resist the safety auditing and runtime monitoring that deployment requires. Existing approaches operate p
Industrial recommender systems commonly use cascaded retrieval, ranking, and re-ranking pipelines. Although efficient, these pipelines fragment information and objectives across modules, rely on rigid rules, and have lim
Morphological transforms are long-standing tools for shape and mask processing, but the de facto reference implementation in the Python ecosystem, i.e. scipy.ndimage, is CPU-only, single-array, and therefore unusable ins
Machine translation tests masked diffusion language models (dLLMs) because every source token must be rendered faithfully, while fixed canvas decoding must choose target length before denoising. Existing masked diffusion
Procedural video-language models must solve heterogeneous tasks from the same visual evidence, including action recognition, forecasting, and procedure prediction. Dense transformer decoders share the same feed-forward n