October 27, 2025 | 6 minute read
Algoverse at NeurIPS 2025: 60 Unique Papers by 230 Students
Hey, Kevin from Algoverse AI Research here!
I’m beyond thrilled to share our most significant milestone to date: so far, 60 unique papers authored by 230 Algoverse students have been accepted to NeurIPS 2025 in San Diego, including several Spotlights, with more expected to follow. If you’re new to the space, Neural Information Processing Systems (NeurIPS) is widely recognized as the most prestigious conference in artificial intelligence and machine learning. Publications at NeurIPS represent groundbreaking contributions and are commonly associated with leading universities and industry leaders like Google DeepMind.
For context: we went from 7 NeurIPS papers last year to 60 this year. That growth has been the result of deeper mentorship, tighter processes, and higher standards this past year.
And we’ve seen what this momentum does for students. Last year’s authors went on to Stanford, Berkeley EECS, CMU, and other top programs. Additionally, many turned their projects into research roles and AI/ML internships.
The ongoing list of this year’s NeurIPS acceptances is listed below. If you want to do serious research with a team that will push you and stand behind your work, you’re in the right place, and we encourage you to join our upcoming 2025 Winter Cohort on November 2.
Accepted Papers
AI for Health
- [Spotlight] Examining the Vulnerability of Multi-Agent Medical Systems to Human Interventions for Clinical Reasoning – Dillon Mehta, Rishi Malhotra, Adam Zobian, Yong Ying Tan, Samir Chopra, Daniella Rand, Natalie Pang, Abhiram Gudimella, Raghav Thallapragada, Derek Jiu, Prisha Shah
- Multi-Turn LLM Systems for Diagnostic Decision-Making: Considerations, Biases, and Challenges – Sejong Kim, Drona Thoka, Varun Puttagunta, Kaylin Sheng, Mark Li, Adnan Ahmed, Thi Uyen Hanh Le, Sai Chidvilas Gudiboina, Ali Ugur
ARLET
- Idea: Fairness Constraints as Reliability Guarantees for RLHF Reward Models – Advay Samnerkar, Doelle Bhattacharya
BioSafe GenAI
- Prompting Toxicity: Analyzing Biosafety Risks in Genomic Language Models – Akshay Murthy, Mengmeng Zhang, Shanmukhi Kannamangalam
- Weight- and Activation-Based Steering Methods Provide Complementary Control of Biochemical Traits in Protein Language Models – Armaity Katki, Nathan Choi, Son Sophak Otra
CCFM
- When Less is More: 8-bit Quantization Improves Continual Learning in Large Language Models – Michael Shihong Zhang, Rishi Adi Ruia, Arnav Kewalram, Saathvik Dharmapuram
Cognitive Interpretability
- A Few Bad Neurons: Isolating and Surgically Correcting Sycophancy – Claire O’Brien, Jessica Seto, Dristi Roy, Aditya Dwivedi
- DecepBench: Benchmarking Multimodal Deception Detection – Vittesh Maganti, Nysa Lalye, Ethan Braverman
DL4C
- DuoLens: A Framework for Robust Detection of Machine-Generated Multilingual Text and Code – Shriyansh Agrawal, Aidan Lau, Sanyam Shah
Entity Reasoning
- Chopping Trees: Semantic Similarity Based Dynamic Pruning for Tree-of-Thought Reasoning – Xirui Huang
- Active Inference Control: Steering, Not Just Scaling, Language Model Reasoning – Josh Karthikeyan, Kai Fu, Derek Jiu
- SwiftSolve: A Self-Iterative, Complexity-Aware Multi-Agent Framework for Competitive Programming – Adhyayan Veer Singh, Aaron Shen, Brian Law, Ahmed Ismail
- Extending AutoCompressors via Surprisal-Based Dynamic Segmentation – Srivishnu Ramamurthi, Richard Xu, Raine Ma, Dawson Park, David Guo
- Inference-Time Chain-of-Thought Pruning with Latent Informativeness Signals – Sophie Li, Nicholas Huang, Nina Luo, Vincent Lin
- LoRA-Guided PPO for Cost-Aware and Compute-Efficient Agent Orchestration – Aneesh Durai, Joshua Cong Hu, Kevaan Buch
- Confidence-Coverage Gating for Early Exit – Aaroosh Rustagi, Hsien Xin Peng, Khushal Murthy, Attrey Koul
Fairness and Reliability in Language Models
- FRIT: Using Causal Importance to Improve Chain-of-Thought Faithfulness – Anand Swaroop, Akshat Nallani, Saksham Uboweja, Adiliia Uzdenova, Michael Nguyen
- Peek-a-Boo Reasoning: Contrastive Region Masking in MLLMs – Anjana Nair, Yushen Li, Adhitya Rajendra Kumar
- Scratchpad Thinking: Alternation Between Storage and Computation in Latent Reasoning Models – Sayam Goyal, Brad Peters, María Emilia Granda, Akshath Vijayakumar Narmadha, Dharunish Yugeswardeenoo
- Limits of Emergent Reasoning of Large Language Models in Agentic Frameworks for Deterministic Games – Chris Su, Harrison Li, Matheus Marques
FM4LS (Foundations and Mechanistic Interpretability for Life Sciences)
- Mechanistic Interpretability of Semantic Abstraction in Biomedical Texts – Nikhil Gourisetty, Snata Mohanty, Vishnu Srinivas, Soumil Jain
Generalization and Process Control in Generation
- Adaptive Originality Filtering: Rejection-Based Prompting and RiddleScore for Culturally Grounded Multilingual Riddle Generation – Duy Le, Kent Ziti, Evan Girard-Sun
- Emotional Framing as a Control Channel: Effects of Prompt Valence on LLM Performance – Enmanuel Felix-Pena, Tiki Li, Wayne Chen, Ethan Hin
Imageomics
- Novel Finetuning Strategies for Adapting Biomedical Vision Language Models to Organ-Centered Pathology Microscopy Tasks – Siddharth Venkatesh, Ayman Sheikh, Anne Essien Essien, Pratibh, Rayhan Roswendi, Jeremiah Zhang
Language Agents Workshop
- AgentChangeBench: A Multi-Dimensional Evaluation Framework for Goal-Shift Robustness in Conversational AI – Manik Rana, Calissa Man, Jeffrey Paine, Anotida Expected Msiiwa
LLM Evaluation
- Adversarial Behavior in Research Settings: Conducting Control Evaluations with RE-Bench – Harini Rajakumar, Vanessa Nwauwa
- GASLIGHTBENCH: Quantifying LLM Susceptibility to Social Prompting – Lening Nick Cui, Sahil Ghosh, Gareth Lee, Xuanzhe Yao, William H. Logian, Michael Li, Ellie Podoshev
- Extending AutoCompressors via Surprisal-Based Dynamic Segmentation – Srivishnu Ramamurthi, Richard Xu, Raine Ma, Dawson Park, David Guo
- GUARD: Guiding Unbiased Alignment through Reward Debiasing – Advay Samnerkar, Doelle Bhattacharya
- MISCHIEF: A Benchmark in Minimal-Pairs of Safety and Culture for Holistic Evaluation of Fine-Grained Image-Caption Alignment – Sagarika Banerjee, Tangatar Madi, Advait Swaminathan, Nguyen Dao Manh Anh
- Predicting Emergent Software Engineering Capabilities by Fine-tuning – Terry Huang, Henry Velasquez
- When Less is More: 8-bit Quantization Improves Continual Learning in Large Language Models – Michael Shihong Zhang, Rishi Adi Ruia, Arnav Kewalram, Saathvik Dharmapuram
LockLLM
- AutoAdv: Automated Adversarial Prompting for Multi-Turn Jailbreaking of Large Language Models – Aashray Reddy, Andrew Zagula, Nicholas Saban
- LSMAS (LLM Security Modeling via Activation Steering) – Anthony Kuang, Ahmed Ismail
- User Confidence-Fueled Stereotypes: Investigating Sycophantic Amplification of Implicit Bias in Language Models – Hannah You, Daniel Wang, Victor Chan, Mirabel Wang
Mechanistic Interpretability
- [Spotlight] Shared Parameter Subspaces and Cross-Task Linearity in Emergently Misaligned Behaviour – Daniel Aarao Reis Arturi, Eric Zhang, Andrew Adrian Ansah
- [Spotlight] Scratchpad Thinking: Alternation Between Storage and Computation in Latent Reasoning Models – Brad Peters, Sayam Goyal, María Emilia Granda, Akshath Vijayakumar Narmadha, Dharunish Yugeswardeenoo
- Death by a Thousand Directions: Exploring the Geometry of Harmfulness in LLMs through Subconcept Probing – Saleena Angeline Sartawita, McNair Shah, Adhitya Rajendra Kumar, Naitik Chheda
- Discovering Transformer Circuits via a Hybrid Attribution and Pruning Framework – Hao Gu, Vibhas Nair, Amrithaa Ashok Kumar
- Emergent World Beliefs: Exploring Transformers in Stochastic Games – Michael Ma, Adam Kamel, Tanish Rastogi
- Mitigating Sycophancy in Language Models via Sparse Activation Fusion and Multi-Layer Activation Steering – Pyae Phoo Min, Avigya Paudel, Naufal Adityo, Arthur Zhu, Andrew Rufail
- Universal Neurons in GPT-2: Emergence, Persistence, and Functional Impact – Cheng-Ting Chou, Amrit Kurakula
- What Do Refusal Tokens Learn? Fine-Grained Representations and Evidence for Downstream Steering – Rishab Alagharu, Ishneet Sukhvinder Singh, Anjali Batta, Jaelyn S. Liang, Shaibi Shamsudeen, Arnav Sheth
Mathematics of AI
- Amortized Latent Steering: Low-Cost Alternative to Test-Time Optimization – Nathan Egbuna, Saatvik Gaur
Multi-Turn and Interactive Language Model Systems
- [Spotlight] WOLF: Werewolf-based Observations for LLM Deception and Falsehoods – Mrinal Agarwal, Saad Rana, Theo Sundoro, Hermela Berhe
- SENTINEL: Sentiment Evolution and Narrative Tracking in Extended LLM Interactions – Pranav Anuraag, Ethan Xu, Asher Nerenberg, Alexander Arutchev
- Modeling and Predicting Multi-Turn Answer Instability in Large Language Models – Jiahang He, Rishi Ramachandran, Neel Ramachandran, Aryan Katakam
- SMAGDi: Socratic Multi Agent Interaction Graph Distillation for Efficient High Accuracy Reasoning – Aayush Aluru, Myra N. Malik, Samarth Patankar
Regulation and Responsible ML
- StealthEval: A Probe-Rewrite-Evaluate Workflow for Reliable Benchmarks – Lang Xiong, Nishant Bhargava, Jeremy Chang, Jianhang Hong
Reliable ML
- A Few Bad Neurons: Isolating and Surgically Correcting Sycophancy – Claire O’Brien, Jessica Seto, Dristi Roy, Aditya Dwivedi
- ERGO: Entropy-guided Resetting for Generation Optimization in Multi-turn Language Models – Haziq Mohammad Khalid, Athikash Jeyaganthan, Timothy Do
- StealthEval: A Probe-Rewrite-Evaluate Workflow for Reliable Benchmarks – Lang Xiong, Nishant Bhargava, Jeremy Chang, Jianhang Hong
- Automated Generation of Multilingual Jailbreak Prompts – Jonathan Ding, Khanak Jain, Dhruv Nair
- GUARD: Guiding Unbiased Alignment through Reward Debiasing – Advay Samnerkar, Doelle Bhattacharya
- Cross-Lingual Multimodal Retrieval-Augmented Generation for Open Question Answering in Tamil and Yoruba – Mobareji Abejide, Arya Ram
Safe and Explainable Agents
- [Spotlight] Examining the Vulnerability of Multi-Agent Medical Systems to Human Interventions for Clinical Reasoning – Dillon Mehta, Rishi Malhotra, Adam Zobian, Yong Ying Tan, Samir Chopra, Daniella Rand, Natalie Pang, Abhiram Gudimella, Raghav Thallapragada, Derek Jiu
- Automated Specialization of Stateful Agent Systems – Myan Vu, Harrish Ayyanar, Pang Jiang, Anwiketh Reddy
Spatial Vision and Language Environments
- Grounding Foundational Vision Models with 3D Human Poses for Robust Action Recognition – Nicholas Babey, Tiffany Gu, Yiheng Li
- COREVQA: Spatial Reasoning and Multi-Step Visual Entailment in Crowded Environments – Kazuma Choji, Ishant Yunay Chintapatla, Naaisha Agarwal, Andrew Lwin
Speech, Image, and Multimodal Generation Models
- Cross-Lingual Multimodal Retrieval-Augmented Generation for Open Question Answering in Tamil and Yoruba – Mobareji Abejide, Arya Ram
Universal Representations
- [Oral] Shared Parameter Subspaces and Cross-Task Linearity in Emergently Misaligned Behaviour – Daniel Aarao Reis Arturi, Eric Zhang, Andrew Adrian Ansah
- Task Matrices: Linear Maps for Cross-model Finetuning Transfer – Darrin O’Brien, Dhikshith Gajulapalli, Alexander Ramsey, Rishi Nalem
Urban AI Systems
- Enhancing Rural Autonomous Driving Performance with Diffusion-Augmented Synthetic Datasets – Siddharth Arun, Trisha Panchangmath, Saanvi Celamkoti, Vayden Wong
Seeing Algoverse students present alongside graduate researchers and industry leaders at NeurIPS 2025 underscores what’s possible when talented students receive direct mentorship and resources to pursue serious research. With 60 unique papers already accepted, and more decisions still to come, this year stands as a testament to the potential of young researchers pushing the boundaries of modern AI.