FazBrowse GitHub Viewer | Trending |
URL:
| Home
Tools: [Original HTTPS Page]

alphadl (Liam Liang Ding) · GitHub

Overview Repositories 69 Projects Packages Stars 1.6k
🎯
hiring @ alibaba https://liamding.cc/hiring.html
🎯
hiring @ alibaba https://liamding.cc/hiring.html
266 followers · 221 following

Achievements

x2x2

Achievements

x2x2

Highlights

  • Pro

Block or report alphadl

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Close all issues, pull requests, and discussions opened by this user Content in all repositories owned by your account will be closed.
Add an optional note
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
alphadl/README.md

Hi there

🙋‍♂️ I am building a deterministic agentic AI ecosystem at Alibaba. I was the chief scientist at a startup (raised more than 50M$), previously worked at JD Explore Academy and Tencent AI Lab, and held an adjunct researcher position at ZJU.

🔭 Working on the whole pipeline of LLM R&D and their human-centric applications, including efficient and sufficient training, alignment, evaluations, compression, multilinguality, multimodality, agentic application, and much more.

💪 I'm keen on bodybuilding (5 years+), marathon (completed first half marathon (126min) in Beijing-2016 and most recent half marathon (86min) in Sydney-2019😅. will resume training in 2024💪🏻).

🥗 I (once😅) enjoy cooking.

🐈 I like to spend Sundays with my cats (two from 2020-2023, one from 2023).

🔥 Recent open-source projects — agentic AI (data, evaluation, context) and LLM alignment / policy optimization:

  • 🔄 AgentHER Hindsight relabeling of failed trajectories for training.
  • 🧬 AgentSynth Synthetic agent data from scratch with execution validation.
  • 📏 AdaRubric Dynamic rubric evaluation for trajectory quality.
  • 🗜️ trajectory_tokenization ReAct with compressed history for long-horizon context.
  • 📡 SigFibPO SNR-calibrated trust regions and causal fiber residuals for multi-domain RLVR (research code + verl hook).

Pinned Loading

  1. THUNLP-MT/MT-Reading-List THUNLP-MT/MT-Reading-List Public

    A machine translation reading list maintained by Tsinghua Natural Language Processing Group

    TeX 2.4k 440

  2. lookahead.pytorch lookahead.pytorch Public

    lookahead optimizer (Lookahead Optimizer: k steps forward, 1 step back) for pytorch

    Python 336 62

  3. AgentHER AgentHER Public

    AgentHER: Hindsight Experience Replay for LLM Agents

    Python 94 11

  4. AdaRubrics AdaRubrics Public

    AdaRubric: Adaptive Dynamic Rubric Evaluator for Agent Trajectories

    Python 356 37

  5. darts.pytorch1.1 darts.pytorch1.1 Public

    Implementation with latest PyTorch (v1.1) for multi-gpu differentiable architecture search https://arxiv.org/abs/1806.09055

    Python 83 28

  6. 3d-gen-for-llm-builders 3d-gen-for-llm-builders Public

    A hands-on guide to 3D latent diffusion for LLM/VLM builders

    Shell 27

Back | FazBrowse Home | New Git URL