SURL AI News
繁中English日本語

Archive

最新模型

新模型、開放權重、多模態、語音、影像與模型架構。

GitHub Repo

MiniCPM5-2B Opens Training Data and On-Device Deployment Formats, but Quantized-Version Sampling Defaults Draw Scrutiny

OpenBMB has released a 2B dense model with a 131K-token context window and tool-calling capabilities, along with staged checkpoints, training data, and multiple quantization formats. Third-party evaluations affirm its competitiveness within its class, but community testing suggests the official sampling settings may cause GGUF quantizations to fall into repetitive generation loops.
Applied AI research

Hugging Face and Earthmover Connect Aurora and ERA5, Shifting the AI Weather Forecasting Bottleneck to the Data Pipeline

The newly released executable workflow streams ERA5 initial conditions directly, enabling Aurora to generate and backtest forecasts locally or on per-minute-billed compute. It does not introduce a new model; its technical value lies in connecting weights, meteorological arrays, static variables, and validation steps into a reproducible pipeline.
最新模型

Boundary-Aware Self-Distillation Narrows the Scope of LLM False Refusals, but Safety Gains Still Trade Off Against Over-Refusal

A new study models safety policies as fine-grained refusal boundaries within topics and uses paired prompts and compensatory data to prevent models from blocking entire sensitive topics. Experiments on Qwen3-8B show that the method can substantially reduce false refusals near the boundary, but the findings remain limited to political persuasion, a single primary model, and a data pipeline that has not been fully disclosed.
最新模型

IBM Granite Time-Series Model Switches to Conformer and Overlapping Patches, Outputs 99-Quantile Forecasts

PatchTST-FM-r2 combines convolution-assisted attention, an 8,192-step context window, and probabilistic forecasting to target general-purpose, zero-fine-tuning time-series prediction. The model ranks second in GIFT-Eval’s reproducible zero-shot track, but the results were submitted by IBM and still await confirmation through integration into the benchmark project.
推論系統

vLLM 0.29 Makes Model Runner V2 the Default for All Models, With Upgrade Implications for Caching, Communication, and Startup Interfaces

vLLM 0.29.0 completes the transition to Model Runner V2 as the default while expanding speculative decoding, reinforcement learning weight synchronization, and support for multiple hardware backends. This is not a minor update that can be adopted by simply replacing the image: existing deployments must revalidate memory configurations, output consistency, and removed interfaces.
最新模型

NeoHorse-1 Feeds Agent Execution Traces Back into Model Training, but “Recursive Self-Improvement” Remains a Single-Round Prototype

NeoHorse-1 uses a router to collect traces from different models completing agentic tasks, then trains 4B and 9B models through curriculum-based SFT and on-policy distillation. Its official aggregate scores exceed those of same-size Qwen3.5 baselines, but it has yet to demonstrate that repeated rounds of self-updating can deliver sustained gains or prevent data drift.