Tsinghua KEG (THUDM)

#ChatGLM #GLM130B #CodeGeeX #CogVLM #CogView #AMiner The Knowledge Engineering Group (KEG) and THUDM at @Tsinghua_Uni @jietang @ericdongyx

[Tsinghua KEG (THUDM) @thukeg]

#VisualAgentBench: 4o, 4o-mini, 3.5-sonnet currently have an edge as visual foundation agents for now, but open models InternVL & GLM-4V are catching up fast, a similar story to LLMs as agents as revealed in #AgentBench back in Aug 2023.
https://t.co/1LoiPkjHAx
https://t.co/ddbELDos0T

[Tsinghua University @Tsinghua_Uni]

Prof. Liu's team built an #AI doctor for everyday #healthcare! In a #virtual hospital, it treated 10K+ virtual patients with 93% accuracy. They covered 300+ diseases across 21 departments & released BioMedGPT, PathOrchestra, and more for a full #medical AI pipeline. #THUAndBeyond

[Tsinghua CS @thudcst]

🏆Congrats to the Storage Research Group from #Tsinghua DCST for winning the#ASPLOS2025/#EuroSys2025 Large-Scale Model Inference Optimization Contest in Rotterdam! They outperformed global competitors, boosting inference performance by 1.1x using AWS NKI framework optimizations.

[Stanford AI Lab @StanfordAILab]

Check out our latest blog post about MiniVLA, a smaller open-source vision-language-action model!
https://t.co/6TvEMWGBSF

[Paul Vicol @PaulVicol]

Ruslan Salakhutdinov at the Adaptive Foundation Models Workshop!

[Richard Socher @RichardSocher]

AI has a "last-mile problem" similar to self-driving cars.

With self-driving cars, early demos impressed, but real-world deployment took years.

It's easy to hack up a prototype, but making it work reliably at scale is hard.

If each step of an AI agent is only 95% accurate, none of the 30-step workflows will work reliably.

Going from 95% to 99.9% accuracy is the real challenge.

[Z.ai @Zai_org]

🌈AndroidLab: a comprehensive platform for developing and evaluating Android agents.

By integrating a controlled environment and standardized benchmarks, and leveraging the Android Instruct dataset, we significantly boost open-source model performance.

https://t.co/qLac9Rjvbq

[Yuxiao Dong @ericdongyx]

#AutoGLM: Autonomous Foundation Agents for GUIs by @ShawLiu12 and team at @thukeg & @ChatGLM!

Here are some AutoGLM for phone use demos --- beta testing since Oct 25 --- and its tech report
https://t.co/ONkwT5rllu

[Z.ai @Zai_org]

Thank you to the passionate developers for your continued support and patience. CogVideoX-5B-I2V, release!😀

Github: https://t.co/VNpl283CPS

CogVideoX-5B-I2V model: https://t.co/85AiDO6YcD

Gradio space: https://t.co/f0dR1IqrCT

[Tiezhen WANG @Xianbao_QIAN]

What has just happened? @thukeg has just released the CogVideoX image-to-video generation model.

Amazing result.

Combined demo of T2V/I2V and V2V: https://t.co/HgmFRUc1QM

Please duplicate the space with a L4s to avoid the long waiting queue.

Model: https://t.co/3gxwGRicsu

[Gradio @Gradio]

LongWriter-glm4-9b from @thukeg is capable of generating 10,000+ words at once!🚀

Paper identifies a problem with current long context LLMs -- they can process inputs up to 100,000 tokens, yet struggle to generate outputs exceeding lengths of 2,000 words.

Paper proposes that an LLM's effective generation length is inherently bounded by the sample it has seen during supervised fine-tuning😮

Demonstrates that existing long context LLMs already possess the potential for a larger output window--all you need is data with extended output during model alignment to unlock this capability.

Code & models are released under Apache License 2.0🧡

[AK @_akhaliq]

New from @thukeg

LongWriter: Unleashing 10,000+ Word Generation from Long Context LLMs

author @realYushiBai is active in discussion section to answer your questions: https://t.co/UeebckjJjf

[Yushi Bai @realYushiBai]

Thanks @_akhaliq! We find that your long context LLM is secretly a LongWriter💡All you need is data with extended output during model alignment to unlock this capability. Our code, data, and models: https://t.co/9KN9fKWFLC

[AK @_akhaliq]

LongWriter

Unleashing 10,000+ Word Generation from Long Context LLMs

discuss: https://t.co/CHHl12U7RF

Current long context large language models (LLMs) can process inputs up to 100,000 tokens, yet struggle to generate outputs exceeding even a modest length of 2,000 words. Through controlled experiments, we find that the model's effective generation length is inherently bounded by the sample it has seen during supervised fine-tuning (SFT). In other words, their output limitation is due to the scarcity of long-output examples in existing SFT datasets. To address this, we introduce AgentWrite, an agent-based pipeline that decomposes ultra-long generation tasks into subtasks, enabling off-the-shelf LLMs to generate coherent outputs exceeding 20,000 words. Leveraging AgentWrite, we construct LongWriter-6k, a dataset containing 6,000 SFT data with output lengths ranging from 2k to 32k words. By incorporating this dataset into model training, we successfully scale the output length of existing models to over 10,000 words while maintaining output quality. We also develop LongBench-Write, a comprehensive benchmark for evaluating ultra-long generation capabilities. Our 9B parameter model, further improved through DPO, achieves state-of-the-art performance on this benchmark, surpassing even much larger proprietary models. In general, our work demonstrates that existing long context LLM already possesses the potential for a larger output window--all you need is data with extended output during model alignment to unlock this capability.

[Tsinghua KEG (THUDM) @thukeg]

#VisualAgentBench: proprietary models (4o, 4o-mini, 3.5-sonnet) currently have an edge as visual foundation agents for now, but open models InternVL & GLM-4V are catching up fast, a similar story to LLMs as agents as revealed in #AgentBench back in Aug 2023.

https://t.co/1LoiPkjHAx
https://t.co/ddbELDos0T

[Xiao Liu (Shaw) @ShawLiu12]

🚨Thrilled to present VisualAgentBench (VAB) with @yugu_nlp and Tianjie, where we enable both TRAINING & TESTING of visual foundation agents across 5 different environments!
In all 17 large multimodal models (LMMs) are tested. Find our paper, data, and more insights below 👇

Paper: https://t.co/EtURrhGZe3
Code & Data: https://t.co/XrsG9cJwkp

Thanks @_akhaliq for sharing on today’s arxiv on HF!

[Z.ai @Zai_org]

We are not just doing “demo only” for video generation.
Ying, we are bringing a video generation AI that everyone can use.
Create a 6-second video in just 30 seconds.
Try our new product now.
YING:https://t.co/wH5pQusd7s
https://t.co/Rt3eXXR8qB

[Tsinghua CS @thudcst]

🏆Proud moment for us! Our paper on 'Explicit factor models for explainable recommendation'(https://t.co/uL7CVxkqZk) has won the Test of Time Award at #SIGIR2024, leading the way in 'explainable recommendation' since 2014. Congrats to outstanding THUIR group from #DCST, #Tsinghua

[Z.ai @Zai_org]

🚀 We published a tech report about GLM's Family! ChatGLM: A Family of Large Language Models from GLM-130B to GLM-4 All Tools.

https://t.co/cgCKNtE93r

[Aran Komatsuzaki @arankomatsuzaki]

ChatGLM: A Family of Large Language Models from GLM-130B to GLM-4 All Tools

GLM-4:
- closely rivals GPT-4 on MMLU, MATH, GPQA, etc
- gets close to GPT-4 in instruction following and long context tasks

hf: https://t.co/ZQayrOSdn4
repo: https://t.co/ZtxFCfaWcx
abs: https://t.co/Pem3LX2i0P

[AK @_akhaliq]

ChatGLM

A Family of Large Language Models from GLM-130B to GLM-4 All Tools

We introduce ChatGLM, an evolving family of large language models that we have been developing over time. This report primarily focuses on the GLM-4 language series, which includes GLM-4, GLM-4-Air,