Graduate Researcher @ Stanford AI Lab
LLM Post-Training • Agentic AI • Reinforcement Learning • Wireless ML
I am a researcher at Stanford AI Lab (SAIL)** co-advised by Dr. Emily Fox and Dr. John M. Cioffi, focusing on:
- 🧠 LLM Post-Training & Inference — preference optimization, alignment, and reasoning calibration
- 🤖 Internet of Evolving Agents — emergent, self-organizing multi-agent ecosystems
- 📡 Reinforcement Learning & Wireless ML — dynamic decision-making and optimization
- ⚙️ Adaptive Test-Time Compute for Reasoning Models
- Continuous-Utility Direct Preference Optimization (CU-DPO)
- Active Bayesian Preference Models
- Adaptive inference strategies for reasoning accuracy
- RL for high-diversity generation
- Self-evolving multi-agent systems
- Bayesian reputation and dynamic team formation
- Social graph-based coordination
- Emergent specialization (NeurIPS 2026, In Progress)
- GNN-accelerated SDP solvers
- Neural Gaussian Radio Fields
- RL for non-stationary decision systems
- Channel estimation & optimization
| Year | Title | Venue |
|---|---|---|
| 2026 (In Progress) | Internet of Evolving Agents | NeurIPS 2026 (In Progress) |
| 2026 (Submitted) | Detecting & Removing Sycophancy in LLMs | CoLM 2026 |
| 2026 (Submitted) | Continuous-Utility Direct Preference Optimization | ICML 2026 |
| 2026 (Submitted) | Adaptive Test-Time Compute Strategies | ICML 2026 |
| 2026 (Submitted) | Active Alignment with Bayesian General Preference Models | CoLM 2026 |
| 2026 | GNN for Accelerating Low-Rank SDP Solvers | KDD 2026 |
| 2026 | Neural Gaussian Radio Fields for Channel Estimation | TMLR 2026 |
| 2025 | Structured Prompting for Robust Evaluation | (co-authored paper) |
🔎 Full list available on my website & CV.
- Stanford Graduate Fellowship
- Knight-Hennessy Fellowship Finalist
- IEEE Best Workshop Paper Award
- IEEE FIT Best Main Conference Paper Award
- Area Chair: NeurIPS 2025, ICASSP 2026
- Reviewer: ICML, NeurIPS, ICLR, KDD, AAAI
🗓 Sep 2025 — Technical Program Committee Member & Reviewer @ NeurIPS 2025
🗓 Aug 2025 — Recognized as Exemplary Reviewer (IEEE Wireless Communications Magazine)
🗓 Jul 2025 — Founding Member (IEEE SIG on AI-Driven TN-NTN Networks)
🗓 Jul 2025 — ICML 2025 paper accepted + Student Travel Grant award
🗓 May 2025 — ICC Student Travel Grant + Best Workshop Paper Award
🗓 Jan 2025 — 2 papers accepted @ AAAI 2025
🗓 Dec 2024 — 2 papers accepted @ IEEE ICASSP 2025
🗓 Apr 2024 — Rector’s Gold Medal (Best Undergraduate Thesis)
🗓 Jan 2024 — PhD Admission (Stanford Graduate Fellowship)
- Reinforcement Learning
- LLM Alignment & Preference Optimization
- Bayesian Inference
- Graph Neural Networks
- Dynamic Agent Planning
- CUDA & GPU Acceleration
- Linux, Git, VS Code
- Large-Scale Training Pipelines
I am always open to research collaborations and impactful projects in:
- LLM reasoning and alignment
- Multi-agent systems & meta-learning
- Reinforcement learning for dynamic systems
- Wireless + machine learning innovation
📫 Email: muahmed@stanford.edu
🔗 Portfolio: https://ahmd-mohsin.github.io/
“Building adaptive intelligence for the next generation of AI systems.” 🚀