Pai 的 AI 雷达
今天这份简报把代理工具、知识接入和创作/开发工作流的可用能力，与 GitHub Copilot 模型退役、SecurityAdvisory GraphQL 字段增加、GPT-6/Blackwell/DGX Spark 更新及隐私安全研究放在一起；阅读时可先按“能否形成一个可执行步骤”筛选工具，再把平台变更当作接口与兼容性约束，把硬件和研究当作部署、评测与安全判断的依据。
arXiv:2610.00685v1 Announce Type: new Abstract: With the rapid adoption of large language models (LLMs) and parameter-efficient fine-tuning (PEFT) methods, the risk of backdoor attacks has become more severe. Existing backdoor purification methods typically rely on at least one o
arXiv:2610.02039v1 Announce Type: cross Abstract: Recent years have witnessed the rapid adoption of reinforcement learning (RL) in large language model (LLM) post-training, with substantial gains in mathematical reasoning and code generation. In practical systems, however, policy
arXiv:2610.00562v1 Announce Type: cross Abstract: Longitudinal clinical reasoning requires large language models (LLMs) to identify and integrate relevant evidence distributed across extended patient histories. Although long-context models can process increasingly large amounts o
arXiv:2609.32353v2 Announce Type: replace Abstract: Visual token reduction is an effective way to accelerate multimodal large language models (MLLMs), but performance deteriorates rapidly under extremely low token budgets. Existing work has explored both visual-token selection an
arXiv:2610.00196v1 Announce Type: new Abstract: Multimodal large language models (MLLMs) have shown strong potential for video understanding and caption generation, but their performance may decline in specialized medical imaging domains such as echocardiography. This work introd
arXiv:2610.00675v1 Announce Type: cross Abstract: Evolutionary approaches to LLM-driven discovery often generate new programs from a small set of selected ancestors. This keeps contexts manageable but can omit useful evidence from other experiments, whereas including the full exp
arXiv:2610.00025v1 Announce Type: new Abstract: Agent harnesses increasingly want to run small language models (SLMs) on the microtasks around a frontier large language model (LLM) planner: auto-approving shell commands, writing memory, selecting tools, ranking past turns. We ask
arXiv:2610.00872v1 Announce Type: new Abstract: Long-term memory systems for large language models (LLMs) have gained popularity for extending reasoning capabilities across applications. Current memory systems rely on LLM agents to organize and consolidate memory, resulting in co
arXiv:2610.01497v1 Announce Type: new Abstract: Molecular tumor boards integrate genomic findings, clinical context, and therapeutic evidence to support precision oncology. As AI enters this workflow, a key safety challenge is distinguishing truly unsupported recommendations from
arXiv:2610.00354v1 Announce Type: cross Abstract: AI agents that control wallets read attacker-reachable content, so they can be steered into proposing harmful transactions. The usual last line of defense is a pre-signing check: a static allowlist, an LLM reviewer, or a transacti
arXiv:2610.00979v1 Announce Type: new Abstract: Training a single LLM agent jointly across diverse interactive environments has attracted increasing attention as a route to generalist agents. Existing curriculum and data-selection strategies often allocate training at the environ
arXiv:2610.00400v1 Announce Type: cross Abstract: Multi-turn attacks on agentic systems can compose individually permissible actions into harmful outcomes, challenging defenses that assess actions or states in isolation. We show that such attacks leave a detectable signature in t
arXiv:2610.00838v1 Announce Type: cross Abstract: Agentic reinforcement learning (RL) trains a large language model (LLM) to act over long, multi-step interactions. However, a single localized error can cause task failure, while trajectory-level rewards provide limited guidance f
arXiv:2609.12822v2 Announce Type: replace Abstract: Blinded physician evaluation has been considered by many to be the gold standard for assessing clinical reasoning in large language models (LLMs). This is difficult to scale; thus, prior studies typically rely on small physician
arXiv:2609.23889v3 Announce Type: replace-cross Abstract: Automated kernel vulnerability reproduction is essential for bug triage, patch validation, and regression testing, but still lacks an effective and efficient solution. The core challenge is twofold: a reproducer must first
arXiv:2610.00388v1 Announce Type: cross Abstract: Reinforcement learning enables large language model (LLM) agents to learn multi-step behaviors through interaction with their environments. However, rewards in many interactive tasks reflect only the final outcome, providing limit
arXiv:2610.00054v1 Announce Type: cross Abstract: Reading an LLM judge's verdict from the logits of its first generated token is cheap, requires no generation, and is exactly what constrained decoding and likelihood-scoring evaluation harnesses produce. We show that this readout
arXiv:2610.00372v1 Announce Type: new Abstract: Large language model agents rely on external harnesses to pass information between the model and its environment and to recover from execution errors. Yet recovery is usually judged only by average task success. This hides an import
arXiv:2609.38659v2 Announce Type: replace-cross Abstract: We study multi-armed bandits (MAB) with multiple optimal arms, motivated by the fact that many practical decision making problems admit multiple correct answers. For $K$-armed bandits with $A$ optimal arms, we first provid
arXiv:2610.00972v1 Announce Type: new Abstract: As LLM agents undertake increasingly complex, long-horizon tasks, verifying their outputs becomes increasingly challenging. We study how verification capability can be strengthened with a fixed base model, without access to referenc
Albertsons Cos. is using ChatGPT Enterprise and the OpenAI API to help teams work faster and make grocery shopping easier for millions of customers.
You can now post confidential comments on repository security advisories. Confidential comments are visible only to people with write access to the repository, so you can discuss a report with your team without the reporter or other invited collaborators seeing it. Previously, ev
arXiv:2610.02161v1 Announce Type: cross Abstract: Vision-language models (VLMs) and vision-language-action models (VLAs) have recently driven rapid progress in general-purpose robots, yet most progress has focused on single-robot settings. Extending these capabilities to multi-ro
arXiv:2610.01906v1 Announce Type: new Abstract: Scuba divers are taught to control their depth to avoid rapid ascents and descents, which could result in serious injuries such as gas embolisms and barotrauma. However, many underwater tasks necessitate lateral control, maintaining