Jiangjie Chen (陈江捷) is a final year Ph.D. candidate at Fudan University in the School of Computer Science, Shanghai, China. His interested research topics are mostly around autonomous generative agents, including (but are not limited to):
( Download my resumé . Could be outdated. 😶)
Ph.D. in CS, 2019 - 2024 (estimated)
B.S. in CS (honors), 2014 - 2019
Oct. 2023: Check out our newest pre-print Auction Arena! We explore the intriguing domain of how LLMs navigate the complex and dynamic environment of auctions. We introduce AucArena, a novel simulation environment tailored for assessing LLMs within the unpredictable yet strategic world of auctions. Play with arena demo and see if you can beat AI!
Oct. 2023: Our paper SCAR got accepted to EMNLP 2023 Findings! A nice addition to the analogical reasoning domain! See you in Singapore :).
July 2023: Our paper CoScript got an Outstanding Paper Award in ACL 2023!
May 2023: A pre-print on the knowledge conflict of large language models! See the tweet. Turns out ChatGPT and GPT-4 somehow stick to its own belief, are receptive/gullible to longer, better-formatted, more popular evidence, and follow the herd… All kinds of human-like, dangerous behaviors!
May 2023: Check out two pre-prints on Analogical Reasoning, which extend E-KAR! AnalogyKB is a million-scale analogy KB derived from existing KGs, to enable machines to achieve analogical reasoning skills. SCAR is a new challenge for evaluating LLMs’ structure abduction ability for scientific analogies, which is essential for human-like analogical reasoning.
May 2023: Got two papers about LLMs accepted to the main conference of ACL 2023! The first paper is about analyzing why LLMs fail to generate negative knowledge while being able to recognize them. The other is CoScript, studying how to generate plans under constraints with LLMs. See you in Toronto (hopefully :/)!
Feb. 2023: Presenting VENCE on AAAI 2023!
Nov. 2022: Two papers accepted to AAAI 2023! One is VENCE on correcting factual errors in texts, and the other is NEON on explaining why a statement is false: both focus on solving the tasks without direct supervision. Welcome to check it out!
Oct. 2022: Gave a talk at MSRA.
Oct. 2022: I was awarded with China National Scholarship for Doctoral Students.
Sept. 2022: Just married💕!
Sept. 2022: We officially release a new version of the E-KAR dataset (v1.0 -> v1.1), with a substantially improved English dataset! Over 600 problems and 1,000 explanation texts are manually adjusted, and we are as strict as we can! See more information at the E-KAR project page. Have fun!
July 2022: Talk titled “Right for the Right Reasons: Explainable Reasoning on Analogical Recognition and Fact Verification” (in Chinese).
July 2022: ACT for NAT will be presented at NAACL-HLT 2022.
May 2022: E-KAR will be presented at the Commonsense Representation and Reasoning (CSRR) workshop at ACL 2022, discussions welcomed!
May 2022: E-KAR will be presented at ACL 2022 (virtually) in a poster session, welcome to check it out!
Apr. 2022: Our paper (ACT) on non-autoregressive translation got accepted at NAACL-HLT 2022!
Feb. 2022: Our paper (E-KAR) on analogical reasoning got accepted at ACL 2022 (Findings)!
We propose AucArena to tests LLMs in auctions, showing they can strategize but with variable success, indicating potential for enhancement.
We propose an over-generate-then-filter approach to improve large language models (LLMs) on constrained language planning, and use it to distill a novel constrained language planning dataset, CoScript.
We find that large language models (LLMs) speak too positively about negative commonsense knowledge, which is caused by statistical shortcuts and negation reporting bias from language modeling pre-training.
Fact verification guided iterative factual error correction without the supervision from corretion.
We benchmark knowledge-intensive analogical reasoning with human-annotated explanations.