Xin Li李鑫

Ph.D. student · Nanyang Technological University · advised by Prof. Chau Yuen

Xin Li李鑫

I work on LLM agents — evaluating what they can do, training them, and orchestrating them with tools, verifiers and each other to get real tasks done. Lately, on agents that improve themselves.

I’m looking for research internships in 2027, on LLM agents, post-training, and self-improvement.

Google ScholarGitHubHugging FaceCVupdated Oct 2026Email

Portrait of Xin Li

The research, as one loopSelected publications →

02

Train

How can models learn from verifiers, rewards, and other models?

  • DN-MOPD

    In multi-teacher distillation, normalizing each teacher's feedback scale per domain beats label routing alone.

    Preprint 2026

  • ReMaRk

    When a request has several valid human answers, rewarding them jointly makes RL for LLM recommenders cheaper.

    Preprint 2026

Also: ErrOPD · Re:Form · ICSC · GraphReduce →

03

Orchestrate

How should models, tools, and verifiers work together to complete tasks reliably?

  • DebateLedger

    Distinguishes harmful collapse from useful correction in multi-agent debate.

    NeurIPS 2026

  • SYNC

    Heterogeneous language models pass facts by sending shared word indices; each receiver writes them with its own word directions.

    Preprint 2026

Also: TLVC · LACP →

03 → 01. Every new way of orchestrating models creates new failure modes, which need to be evaluated in turn.

So far, mostly where an answer can be checked:formal verificationmathematicscodeNow, self-improvement beyond checkable answers.

RecentAll news →

  1. Recognized as a Top Reviewer for NeurIPS 2026.
  2. Our paper SIM-D2NN (on onboard terrain classification straight from raw SAR data, using a stacked metasurface as the classifier) was accepted to IEEE Transactions on Signal Processing.
  3. Two of our papers were accepted to the NeurIPS 2026 Evaluations and Datasets Track: WirelessMathBench-XL, a wireless-math benchmark shipped with a rerunnable contamination audit, and DebateLedger, a protocol separating harmful collapse from useful correction in multi-agent LLM debate.
  4. Our paper ICSC (on reconstructing wireless signals for low-altitude networks with small language models trained by reinforcement learning) was published in npj Wireless Technology.
  5. Two of our papers were accepted to EMNLP 2026: TLVC, on picking the right verifier for best-of-K reasoning selection (Findings), and GraphReduce, on coverage-preserving LLM aggregation of e-commerce reviews (Industry Track).

Before the Ph.D.

Robot perception — visual-inertial odometry at MEGVII, RGB-D + IMU indoor mapping at Microsoft Research Asia, and multimodal localization at Gausium Robotics, where I led a five-engineer team and shipped to a fleet of 1,000+ commercial cleaning robots. More →

Contact

Looking for research internships in 2027, and always glad to talk about research or collaboration.

xin019@e.ntu.edu.sgScholarGitHubHugging FaceLinkedInCVupdated Oct 2026