纪焘 (Tao Ji)

Hi, great work! Would it be possible to share the 15K `Russian Syntagrus` treebank obtained by random sampling in the paper? Thanks!

Reproduction: about the specific Contriever model

1

Hi there! Thanks so much for your work on the LaMP benchmark. I've been working on reproducing the results but found myself a bit puzzled about the retriever. Facebook has...

Hello, Thank you for providing the pre-trained dataset for TinyLlama. After downloading the dataset from [tinyllama_pretrained_data](https://huggingface.co/datasets/TinyLlama/tinyllama_pretrained_data), could you give an example or guidance on how to load and use this...

Has InternLM-XComposer2-VL (7B/1.8B) undergone interleaved text-image pretraining? Thanks!

[Feature Request] Flexible MLA Kernel Supporting More `d`, `dv`, and `n_heads` Values

5

## Background Hi TileLang team! Thank you for the fantastic repo — the performance of the [MLA](https://github.com/tile-ai/tilelang/tree/main/examples/deepseek_mla) kernel is truly impressive. We recently proposed a framework called [MHA2MLA](https://github.com/JT-Ushio/MHA2MLA), which enables...