最新模型

新模型、開放權重、多模態、語音、影像與模型架構。

SafeKeep Finds Tool Schemas Weaken Models’ Refusal Signals, Cutting Prompt Injection Success to 2.5%

New research suggests that the JSON-style tool specifications used by agents are not merely neutral interfaces and may suppress models’ internal refusal representations. By separately evaluating safety before execution using flattened text specifications, SafeKeep increased the average refusal rate for harmful requests from 23.8% to 70.6%.

2026-08-03

AI 基礎設施

Etched Unveils Rack-Scale Inference Architecture, Challenging GPUs with Low-Voltage Compute and Cluster-Shared Memory

Etched’s first N4P inference chips have entered rack-level validation. The system uses low-voltage compute arrays and cluster-scale memory to address throughput and latency bottlenecks, respectively. The company claims sparse MoE workloads can sustain more than 80% of peak FLOPs, but has yet to publish reproducible model, power, or end-to-end performance data.

2026-08-03

代理工具

Echoverse Trains Computer Agents with Evolvable Applications, Raising the 9B Model’s Cross-Test Success Rate to 67.1%

A Microsoft Research team jointly updates synthetic applications, tasks, and graders based on agent failure trajectories instead of merely adding more static webpages. The 9B model’s average success rate across 14 evaluations rose from 36.5% to 67.1%, although only four environments have been publicly released as benchmarks.

2026-08-03

代理工具

Qwen-UI-Agent Unifies Mobile, Desktop, and Browser Operations, With Online RL Trajectories Exceeding 100 Turns

Alibaba’s Qwen team has released a technical report on a GUI agent spanning mobile, desktop, web, and DeepSearch environments. Its unified action space combines on-screen interactions, CLI commands, and batched actions. The model leads existing systems on several self-reported benchmarks, but its weights, training data, and complete execution framework have not yet been released.

2026-08-03

代理工具

OSReward Evaluates 27 Computer-Agent Judges, With Hard Trajectories Exposing a Widespread “False Success” Bias

Using 1,019 human-annotated trajectories spanning the web, Windows, Ubuntu, and mobile platforms, OSReward tests whether vision-language models can correctly determine whether computer agents have completed their tasks. The team also released 100,000 reasoning-based judgment examples and trained self-hostable 9B and 35B OS-Shepherd reward models.

2026-08-02

最新模型

Cosmos 3 Edge Shrinks the World Model to 4B, Handling Vision, Text, and Robot Actions on a Single GPU

NVIDIA has released the Cosmos 3 Edge weights and inference pipeline, combining autoregressive reasoning and diffusion-based multimodal generation in a 4B Mixture-of-Transformers model. The model accepts text, images, and action trajectories, and a robot policy variant fine-tuned on DROID data is also available, though physical consistency and edge-device latency still require independent verification.

2026-07-23

AI 安全

AISPA 稽核商用 AI 的隱藏系統提示，約四成產品含損害使用者利益的指令

AISPA 把系統提示拆成八項使用者保障維度，並分析 88 款商用 AI 產品中的 3,249 條指令。雖然 98.9% 的產品至少包含一項保護性規則，完整覆蓋八個面向的比例只有 24%。

2026-08-02

AI 安全

OWASP Agentic Skills Top 10 Completes Public Review for v1, Treating the Skill Supply Chain as a Distinct Attack Surface

The public review of the OWASP Agentic Skills Top 10 v1 draft has concluded. The framework organizes malicious skills, excessive permissions, dependency confusion, and cross-platform portability risks into ten control categories. It treats skills as the behavioral layer between models and MCP tools and proposes a universal format for declaring permissions, provenance, and integrity.

2026-08-02

最新模型

POCKET Compresses a 35B Sparse MoE to 8.2 GB, Runs on Phones and CPU-Only Systems With Standard llama.cpp

VIDRAFT has released the weights for its POCKET series, quantizing a 35B model down to as little as 8.2 GB while retaining roughly 3B active parameters per token. The models run directly with upstream llama.cpp, although the compression method remains undisclosed and current performance and quality evidence comes primarily from the development team.

2026-08-02

AI 安全

Google Earth Pulled Its Image Generation Feature After One Day as SynthID Failed to Stop Fake Satellite Images From Spreading

After Google connected Nano Banana 2 to real-world geospatial imagery, users quickly generated and shared screenshots of fabricated explosions, airstrikes, and other scenes. The company withdrew the feature the next day. Although generated images carried watermarks and were not added to shared maps, the incident shows that provenance labels cannot replace input restrictions and interface isolation.

2026-08-02

最新模型

openPangu 2.0 Pro Releases Open 505B MoE Weights, Supports a 512K Context with 18B Active Parameters

Huawei’s openPangu team has released a 505B mixture-of-experts model trained on Ascend NPUs, activating approximately 18B parameters per token. The model introduces hierarchical sparse attention and three-head multi-token prediction, though its performance figures currently come primarily from the team’s own evaluations.

2026-08-02

科學機器學習

UNICON Unifies Nine Classes of Numerical Systems With 41.6 Million Parameters, Enabling Cross-Domain Forecasting With Frozen Weights

UNICON converts data from rivers, traffic networks, power grids, weather systems, and human mobility into a shared graph-sequence interface, using a small number of numerical examples to specify new tasks in context. In unseen domains such as web traffic, the model approaches the performance of specialized models, but it has so far been validated only on forecasting tasks, and neither its weights nor code have been released.

2026-08-02