FangruLin99 Profile Banner
Fangru Lin Profile
Fangru Lin

@FangruLin99

Followers
4K
Following
271
Media
36
Statuses
163

Research Intern @GoogleDeepMind; DPhil student @UniofOxford; Clarendon Scholar; Prev @MSFTResearch, @Microsoft, @turinginst; Computational Linguist

Mountain View, CA
Joined January 2024
Don't wanna be here? Send us removal request.
@FangruLin99
Fangru Lin
21 days
✨Our recent work is covered by @guardian! Honored to have contributed to the project and worked with all fantastic people!
@oiioxford
Oxford Internet Institute
22 days
“Benchmarks underpin nearly all claims about advances in AI,” Andrew Bean @oiioxford told the @guardian. “But without shared definitions and sound measurement, it becomes hard to know whether models are genuinely improving or just appearing to.”
0
1
14
@FangruLin99
Fangru Lin
3 months
I have started my research internship @GoogleDeepMind in Gemini Deep Research team with Kai Zhao and @v_mudit! DM me if you are around Mountain View and want to get a coffee! Super excited about the new journey!🥳
51
21
1K
@lawhy_X
Yuan He
3 months
‼️We received several reports that OpenReview is currently down. We will extend the deadline to the end of the day when the system gets resumed.
@lawhy_X
Yuan He
4 months
🚨 [Call for Papers] SEA @ NeurIPS 2025 🚨 Scaling Environments for Agents (SEA) Workshop 📅 December 6, 2025 | 📍 San Diego, USA We're excited to invite submissions to the SEA Workshop at NeurIPS 2025! 🧵1/n
2
2
15
@FangruLin99
Fangru Lin
3 months
Proud to have contributed 🙌 This project’s neuro-symbolic core is next-level — and Loong is the most cross-domain CoT dataset I know 🔥
@guohao_li
Guohao Li 🐫
3 months
Sir, we built this. A RL environment for learning reasoning at scale. GitHub: https://t.co/KAE1xce6wh HF dataset: https://t.co/9CMpbttSIR We extracted seed datasets from sources like textbooks, code libraries like sympy, networkX, Gurobi (math programming lib), rdkit
0
0
25
@FangruLin99
Fangru Lin
3 months
Now our paper is accepted at @emnlpmeeting 2025 main! 🎉🔥📚 Can’t wait to share more soon 🤩
@FangruLin99
Fangru Lin
4 months
Thanks for talking about our work! 🥳We have released 🗓️ TCP: a natural language conversation benchmark for constraint-based planning. We try to talk about *real-life and diverse* planning tasks! Really proud to be part of the team!🤩🤩🤩
0
1
18
@FangruLin99
Fangru Lin
3 months
🚨 Deadline Extended! 🚨 Our Scaling Environments for Agents 🧑‍💻🤖 workshop at @NeurIPSConf 2025 is still open for submissions! If you’re working on scaling, environments, or agents, we’d love to see your papers! 📄✨ 📅 New deadline: Sept 1st 🧵More information and submit:
@lawhy_X
Yuan He
4 months
🚨 [Call for Papers] SEA @ NeurIPS 2025 🚨 Scaling Environments for Agents (SEA) Workshop 📅 December 6, 2025 | 📍 San Diego, USA We're excited to invite submissions to the SEA Workshop at NeurIPS 2025! 🧵1/n
0
6
17
@FangruLin99
Fangru Lin
4 months
🚨🚨🚨Happening *NOW*!!! Catch us at #ACL2025 Poster session 1 Hall X5-91!!!
@FangruLin99
Fangru Lin
4 months
I will be presenting our paper on dialect fairness&robustness of reasoning with @iperboreo_ and @vjhofmann Monday 11:00-12:30 at Hall 4/5 (poster session 1)! Find us by the pink poster!!!😆😆😆
5
1
13
@FangruLin99
Fangru Lin
4 months
I will be presenting our paper on dialect fairness&robustness of reasoning with @iperboreo_ and @vjhofmann Monday 11:00-12:30 at Hall 4/5 (poster session 1)! Find us by the pink poster!!!😆😆😆
4
3
45
@FangruLin99
Fangru Lin
4 months
Join our workshop at #NeurIPS2025 with our fantastic lineup of speakers! Call for papers and reviewers *NOW*!!!🤩
@guohao_li
Guohao Li 🐫
4 months
🚨 [Call for Papers] SEA Workshop @ NeurIPS 2025 🚨 📅 December 6, 2025 | 📍 San Diego, USA 🌐: https://t.co/ISaakkESxO Environments are the "data" for training agents, which is largely missing in the open source ecosystem. We are hosting Scaling Environments for Agents (SEA)
0
3
31
@FangruLin99
Fangru Lin
4 months
I will attend #ACL2025NLP 🇦🇹! Super excited to be back to Vienna for my first time ACL🤩🤩🤩!! I will be in Vienna from this Saturday onwards. DM me if you want to talk about anything about LLM, especially in planning and reasoning (or just want to explore the city)!
1
1
45
@FangruLin99
Fangru Lin
4 months
🤩 We are hosting a workshop on Scaling Environments for Agents at @NeurIPSConf and we are accepting papers *NOW*!!!
@lawhy_X
Yuan He
4 months
🚨 [Call for Papers] SEA @ NeurIPS 2025 🚨 Scaling Environments for Agents (SEA) Workshop 📅 December 6, 2025 | 📍 San Diego, USA We're excited to invite submissions to the SEA Workshop at NeurIPS 2025! 🧵1/n
0
0
12
@FangruLin99
Fangru Lin
4 months
You might also be interested in this ICML paper if you are interested in constraint-based natural language planning! I will attend ACL next week (although presenting a different paper) so please DM me if you are interested in talking about general LLM or planning!🤩
@FangruLin99
Fangru Lin
1 year
💥Our #ICML2024 camera-ready paper Graph-enhanced Large Language Models in Asynchronous Planning is available on arxiv: https://t.co/UximBQapp2! *Off-the-shelf* method *Plan Like a Graph* gives GPT-3.5/4 Pareto improvement on asynchronous planning tasks of all complexities!🧵
0
1
3
@FangruLin99
Fangru Lin
4 months
Thanks for talking about our work! 🥳We have released 🗓️ TCP: a natural language conversation benchmark for constraint-based planning. We try to talk about *real-life and diverse* planning tasks! Really proud to be part of the team!🤩🤩🤩
@Auss_Abbood
auss
4 months
New benchmark❗️ The worst part in large team projects with complex dependencies between deliverables is ... planning. 🧖💬📅↔️📆🗨️🧖‍♀️ Can't LLMs help with that?
1
0
6
@FangruLin99
Fangru Lin
6 months
I’ll present the same paper at #ACL2025 main in Vienna 🤩! The paper link is: https://t.co/s4KMpj5QUw. You can also find our demo here:
@FangruLin99
Fangru Lin
6 months
My internship work at @MSFTResearch on evaluating dialect fairness and robustness in reasoning tasks is accepted at #ACL2025 main! Super excited to have my first ACL publication! Preprint here: https://t.co/3olDhAYiX0. See you in Vienna!
0
0
6
@FangruLin99
Fangru Lin
6 months
It was so much fun presenting at Natural History Museum! Fantastic science talk with bears and dinosaurs (I checked the bear is touchable 🐻!)
6
4
526
@FangruLin99
Fangru Lin
6 months
My internship work at @MSFTResearch on evaluating dialect fairness and robustness in reasoning tasks is accepted at #ACL2025 main! Super excited to have my first ACL publication! Preprint here: https://t.co/3olDhAYiX0. See you in Vienna!
Tweet card summary image
arxiv.org
Language is not monolithic. While benchmarks, including those designed for multiple languages, are often used as proxies to evaluate the performance of Large Language Models (LLMs), they tend to...
@FangruLin99
Fangru Lin
1 year
🤯Do you know AI does not code so well for African American English users? We release ReDial, the first human-written Standardized-African American English parallel reasoning benchmark! Check it out on our demo and test your LLM for dialect robustness! https://t.co/oG5rvtzXxL
5
3
100
@FangruLin99
Fangru Lin
7 months
Paper link is here:
0
0
3
@FangruLin99
Fangru Lin
7 months
I’m presenting our recent work at LLM reasoning and planning workshop today from 11:50 to 13:20 at Garnet 212-213 @iclrconf ! Come and find me with a pink poster if you are interested in dialect robustness in reasoning!
3
1
58
@FangruLin99
Fangru Lin
7 months
I will be attending @iclr_conf and presenting our paper on reasoning fairness and robustness in a dialect ( https://t.co/QlTzjr8oY5) at Workshop on Reasoning and Planning for Large Language Models! DM me if you want to get a coffee!
1
7
75
@FangruLin99
Fangru Lin
9 months
I believe there are many more possible exciting extensions along this line of work of reasoning formalisation and simulation using verifiable approaches! Similarly also see our ICML paper here!
@FangruLin99
Fangru Lin
1 year
💥Our #ICML2024 camera-ready paper Graph-enhanced Large Language Models in Asynchronous Planning is available on arxiv: https://t.co/UximBQapp2! *Off-the-shelf* method *Plan Like a Graph* gives GPT-3.5/4 Pareto improvement on asynchronous planning tasks of all complexities!🧵
0
0
2