ABCDEFGHIJKLMNOPQ
1
TimestampNameEmailAffiliation/RoleWill you be at ICML?How keen are you for strangers to reach out?Are you looking for opportunities? If so, what kind?Are you recruiting people? If so, for what?Are you looking for research collaborators?Research interestsAny papers of your's you'd be excited to discuss?
2
30/06/2026 21:10:16Steven Dillmannstevendi@stanford.eduPhD Student at Stanford UniversityYes - all of itExcited!YesTerminal-Bench Science: https://www.tbench.ai/news/tb-science-announcement
3
30/06/2026 21:11:17Prashant Rajurajuprashant@gmail.comindependent researcher (on the market)NoExcited!Job (industry), Internship / fellowship, PhD, Informal mentorshipInformal menteesYesrepresentational learning, representational geometry, model evaluation, scientific foundation modelshttps://arxiv.org/abs/2604.04155 https://arxiv.org/abs/2604.17698 https://arxiv.org/abs/2601.09173
4
30/06/2026 21:11:32Andrew RufailAndrew.rufail@aucegypt.eduResearcher at Algoverse AINoExcited!Job (industry), Internship / fellowshipMaybeSafety, post training, applied mech interphttps://openreview.net/forum?id=U4tAVkfVqb
5
30/06/2026 21:11:40Valentin NOËLval.noel@proton.meResearch Scientist @DevoteamYes - all of itExcited!Job (industry), Postdoc, Informal mentorshipJob (industry), Internship / fellowshipYesMechInterp, Graph Signal Processing, NeuroAI, Agentic Guardrails, Cyber for AIGeometry of Reason: Spectral Signatures of Valid Mathematical Reasoning
6
30/06/2026 21:12:21Philipp Mondorfp.mondorf@lmu.dePhD Student at LMU MunichYes - all of itExcited!Job (industry), Internship / fellowship, PostdocYesInterpretability, Reasoning, Compositionality
7
30/06/2026 21:12:39Arun Vignesh Malarkkanamalarkk@asu.eduPh.D. Candidate at Arizona State University Yes - all of itExcited!Job (industry), Internship / fellowshipInformal menteesYesInvariance in latent space in Reasoning Tasks, mechanistic Interpretability, Causal Guided Generative AISource: arXiv https://share.google/xlJVFWHCPR7GQZdYW
8
30/06/2026 21:13:22Méloux Maximemaxime.meloux@univ-grenoble-alpes.frPhD Student, Université Grenoble Alpes/CNRSYes - all of itExcited!MaybeCausal abstraction, mech interp grounded in statistical inference (bayesian)"The Dead Salmons of AI interpretability" (https://arxiv.org/abs/2512.18792), "Validing Causal Abstraction on Simulated Complex Systems" (https://openreview.net/forum?id=hpyVULfQjb)
9
30/06/2026 21:14:02Jonathn Changjc3683@cornell.eduPhD student at Cornell UniversityYes - all of itExcited!Job (industry), Internship / fellowship, PostdocInformal menteesYesCausal abstractions, especially applied to circuit tracing / circuit sparsity techniques. (unrelated: character training, evals for value alignment)PLOT: Progressive Localization via Optimal Transport in Neural Causal Abstraction (https://arxiv.org/abs/2605.06979)
10
30/06/2026 21:14:31Philippe Chlenskip@chlenski.comResearch Scientist at MetaYes - all of itExcited!Informal mentorshipInformal menteesMaybeMechanistic interpretability, explainable AI, representation learningSurrogate Fidelity: When Can Open LLMs Explain Closed Ones? https://openreview.net/forum?id=lzhJ7uBaSu
11
30/06/2026 21:18:15Adam Karvonenadam.karvonen@gmail.comAnthropic FellowsYes - all of itExcited!Job (industry)MaybeActivation Oracles / NLAs
12
30/06/2026 21:18:18Yuqi SunYuqisun29@gmail.com Research scientist and founder of Mindoverflow.aiNoExcited!Informal mentorshipJob (industry), Internship / fellowship, Informal menteesYesApplied interp on hypothesis generation / scalable oversight, persona emergence in pre training,Https://arxiv.org/html/2602.05183v2
13
30/06/2026 21:18:19Raffaello Fornasierefornasiere.raffaello@gmail.comAi Safety Researcher at Arcadia ImpactYes - workshops onlyExcited!Job (industry), Internship / fellowship, PhD, Informal mentorshipYesMechanistic Interpretability, Model Organisms, evaluation and situation awareness, introspectionActivation Oracles (https://arxiv.org/pdf/2512.15674)
Large Language Models Often Know When They Are Being Evaluated (https://arxiv.org/abs/2505.23836?)
Evaluating Frontier Models for Stealth and Situational Awareness (https://arxiv.org/abs/2505.01420)
Subliminal Learning (https://arxiv.org/abs/2507.14805)
14
30/06/2026 21:18:34Koshiro Aokiaokikoshiro@akane.waseda.jpMaster's student at Waseda UniversityYes - all of itExcited!Job (industry), Internship / fellowship, PhD, Informal mentorshipYesIntrospection, Metacognition, Relationship between artificial and human intelligenceCrosscoders Identify Shared or Specific Features between the Human Brain and Language Models ( https://openreview.net/forum?id=07WI8MV4iP )
In-Context Neurofeedback: Can Large Language Models Control Their Internal Representations through Privileged Access? ( https://openreview.net/forum?id=3g6jrASugZ )
15
30/06/2026 21:19:33Edward Stevinsonedward.stevinson@hotmail.comPhD student at Imperial College LondonYes - all of itExcited!Job (industry)MaybeFeature geometry & adversarial robustnessAdversarial Attacks Leverage Interference Between Features in Superposition - https://arxiv.org/abs/2510.11709
16
30/06/2026 21:20:10Hugohbuurmei@stanford.eduPhD Student at StanfordYes - all of itExcited!Internship / fellowshipMaybeInterpretability, self-improvement for robotics
17
30/06/2026 21:20:34Greg Frankgreg@moltaicorp.comChief Scientist at MoltAI CorpYes - all of itExcited!Job (industry), Internship / fellowshipMaybeSmall models for edge deployment, agentic red-teaming, safety.https://arxiv.org/abs/2604.04385
18
30/06/2026 21:21:31Zihan Dongpuma122707@gmail.comIndependent Researcher at LenovoNoExcited!Job (industry)YesComputer Use Agent, World model, Benchmark, Dataset and Anything that is impactful to Everyone
19
30/06/2026 21:22:42Sinie van der Benvandersi@ethz.chPhD Student at ETH ZurichNoExcited!Internship / fellowship, Informal mentorshipYesmanipulation of knowledge (steering), universality, anthropomorphism, feature compositionality, circuits, geometric structure
20
30/06/2026 21:24:03Michael Yumichael@raft.bioMATS fellow and startup founder (Raft Bioworks)Yes - workshops onlyExcited!Informal mentorshipInformal menteesMaybeprotein model interpretability, bio agents, biosafety, protein screening, wet labhttps://arxiv.org/abs/2606.10080
21
30/06/2026 21:24:14Junyu Renjunyuren@uchicago.eduPhd Student at the University of ChicagoYes - all of itExcited!Internship / fellowship, Informal mentorshipYesSafety research, mech interp, technical governance
22
30/06/2026 21:24:30Manav Pandeympandey32@gatech.eduSr. Research Engineer @ Amex, M.S Student at Georgia Tech,Yes - all of itExcited!Job (industry), Informal mentorshipInformal menteesYesCircuit tracing/attribution graphs, lying and malintent, mech interp for noisy/self-supervised models, agent societies + mech interphttps://arxiv.org/abs/2604.19117
23
30/06/2026 21:24:49Julian Aßmanncontact@julianassmann.dePhD Student at University of Mannheim and SAPYes - all of it
24
30/06/2026 21:26:13Aniket Deshpandeaniketdeshh@gmail.comUndergraduate at UIUCNoExcited!Internship / fellowship, PhD, Informal mentorshipYesinterpretability, tensors, statistical physics
25
30/06/2026 21:27:37Gustavo Merciergalexmercier@gmail.comMS Student at Northwestern UniversityYes - all of itExcited!Job (industry), Internship / fellowship, PhD, Informal mentorshipMaybeMy previous research has been centered around the use of Sparse Autoencoders and Crosscoders, but I'm currently more interested in seeing how mechanistic interpretability can be used to improve model performance. Technical topics-wise, I'm interested in reinforcement learning and agentic systems, as well as improving long-horizon performance in AI models.
26
30/06/2026 21:27:59David Shu Cheungdavidsc2@illinois.eduMS Student at UIUCYes - workshops onlyExcited!Job (industry), Internship / fellowship, PhDYesPractical, Applied Interpretability https://openreview.net/pdf?id=TVIZyK2AEW
27
30/06/2026 21:28:41Marco Mascorromarco@marcomascorro.comIndependent ResearcherYes - all of itExcited!Informal mentorshipInformal menteesYesInternal representations of emotion, eval awareness, back propagation representations https://arxiv.org/abs/2604.07729
28
30/06/2026 21:31:19Dayamoy Datta Saikatdayamoydatta96@gmail.comMS student at Miami University, Oxford, OhioNoExcited!Internship / fellowshipYesMechanistic Interpretability, AI security, AI safety
29
30/06/2026 21:32:09Nikita Menonnikitamenon2510@gmail.comAI Safety Researcher at Arcadia Impact (LASR Labs)Yes - workshops onlyExcited!Job (industry), Internship / fellowship, Informal mentorshipYesOversight, Evaluations, Interpretability, Dangerous Capabilities and probably many other areas within AI Safety & Alignment.
30
30/06/2026 21:33:36Ayush Warikooayush.warikoo77@gmail.comResearcher at Meta (Industry)NoExcited!Informal mentorshipMaybeMech Interp, Agent Optimization and DistillationSurrogate Fidelity: When Can Open LLMs Explain Closed Ones? https://openreview.net/forum?id=lzhJ7uBaSu
31
30/06/2026 21:38:04Camila Blankcamilab@stanford.eduMATS fellow with Neel Nanda, undergrad at StanfordYes - all of itExcited!Job (industry), PhDMaybePost-training generalizationhttps://arxiv.org/abs/2606.00995
32
30/06/2026 21:39:29Zach Carmichaelzjcarmichael@gmail.comResearch Scientist at MetaNoExcited!Job (industry)MaybeMechanistic interpretability; Explainable AI; Intrinsically interpretable modelinghttps://openaccess.thecvf.com/content/WACV2024/html/Carmichael_Pixel-Grounded_Prototypical_Part_Networks_WACV_2024_paper.html;
33
30/06/2026 21:40:37Aheli Poddarahelipoddar2003@gmail.comAssociate Software Engineer at Red Hat PyTorch EngineeringNoExcited!Internship / fellowship, PhD, Informal mentorshipYesMechanistic interpretability, Causal AI
34
30/06/2026 21:40:52Truong Caitruongcai@alum.mti.eduAshler AI lead researchYes - main conference onlyExcited!Job (industry), Internship / fellowshipJob (industry)YesLLM sterring
35
30/06/2026 21:45:21Luis M. Montoyaluismmonh@gmail.comIndependent ResearcherNoExcited!Job (industry)YesSituational Awareness, Improving monitoring techniques, Pre-training alignment, world models
36
30/06/2026 21:48:13Carla Munteancarlamuntean22@gmail.comUniversity of CambridgeNoExcited!Internship / fellowship, PhD, Informal mentorshipYesMechinterp, Representation Learning, AI for Science
37
30/06/2026 21:50:08Ameen Patelameen.patel.210@gmail.comResearch with Stanford University PhDs and AlgoverseNoExcited!Internship / fellowshipMaybeCircuit tracing, AI Measurement, IRT modeling, Sycophancy, AI Safetyhttps://openreview.net/pdf?id=jHUNetbSbq, https://arxiv.org/pdf/2605.17173
38
30/06/2026 21:51:11Muhammed Ustaomeroglumustaome@andrew.cmu.eduPhD Student at Carnegie Mellon UniversityYes - all of itExcited!Internship / fellowship, Informal mentorshipYesunderstanding deep learning experimentally or theoreticallyRET paper: https://arxiv.org/abs/2605.09294
39
30/06/2026 21:51:24Lakshya Chaudhrylakshchaudhry@gmail.comUndergraduate student at Purdue UniversityYes - workshops onlyExcited!Internship / fellowship, Informal mentorshipYesMech Interp / AI security
40
30/06/2026 21:54:15Trinidad Borrelltriniborrell@gmail.comPhD Student at Sorbonne University Yes - all of itExcited!Job (industry), Internship / fellowship, Informal mentorshipMaybeFoundation Models, NeuroAI, AstroAI, Representational Geometry, Representational Learning, Persona Vectors
41
30/06/2026 21:54:23Ege Erdogane.erdogan@uva.nlPhD @ U of AmsterdamYes - all of itExcited!Internship / fellowshipYesAI for science and mech interp
42
30/06/2026 21:54:38Manoj Acharyamanoja328@gmail.comStaff ML @ exponential.vcYes - all of itExcited!Informal menteesYescontinual leanring; AI safety
43
30/06/2026 21:56:32Kia-Jüng Yangkia-jueng.yang@uni-goettingen.dePhD Student at the University of GöttingenYes - all of itExcited!Job (industry), Internship / fellowship, Informal mentorshipYesAI Safety, Mechanistic Interpretability, Human Motion Modeling, Multimodal Deepfakes (Gen. + Detection)https://arxiv.org/abs/2605.26772
44
30/06/2026 22:14:08Juniper Lovatojlovato@uvm.eduAssistant Professor of Computer Science at the University of VermontNoNeutralPhD students, PostdocsMaybeMechanistic Interpretability, NLP, Complex Systems, Tech Ethicshttps://arxiv.org/abs/2602.23164
45
30/06/2026 22:16:42Juliana Lijulianali@college.harvard.eduStudent & Researcher at Harvard UniversityYes - all of itExcited!Internship / fellowship, Informal mentorshipYesNLAs, weight-space readers, neural geometry, CoT monitorability, GLP paper
46
30/06/2026 22:17:42Daniel Saeedidsaeedi3@gatech.eduPhD Student at the Georgia TechYes - workshops onlyExcited!Internship / fellowshipMaybeMech Interp and Agentic AIhttps://arxiv.org/abs/2602.12026
47
30/06/2026 22:19:15Dennis Weidwei@us.ibm.comResearch Scientist at IBM ResearchYes - all of itExcited!Internship / fellowship, Informal menteesMaybeAttributing LLM outputs/behavior to inputs, previous outputs (thoughts), internal components, etc.Local Causal Attribution of Chain-of-Thought Reasoning (https://arxiv.org/abs/2606.21821)
48
30/06/2026 22:27:51Shashank Yadavyadav.shashank@mayo.eduPostdoc at the Mayo ClinicYes - all of itExcited!Internship / fellowship, PostdocsMaybeFoundation models on Electronic Health Records, Pathology VLMs
49
30/06/2026 22:35:00Wen Xingwx5rice@gmail.comMATS fellow 8 & 10Yes - all of itExcited!Job (industry), Internship / fellowship, PhD, Informal mentorshipYesNLA, PCD, anything related to turning compute into model internal understanding https://transformer-circuits.pub/2026/nla, https://transluce.org/pcd, https://transluce.org/self-explanations
50
30/06/2026 22:36:26Ethan Hadleyekhadley@gmail.comGrad Student at Southern Illinois UniversityYes - all of itExcited!Job (industry), Internship / fellowship, PhD, Postdoc, Informal mentorshipYesDecomposition/reverse engineering, activation explainers, AI control, training dynamics, evaluating interp methodshttps://openreview.net/pdf?id=a2sc2Y91hO
51
30/06/2026 22:43:05Elizabeth Qiuelizqiu@gmail.comindependent researcherNoExcited!Job (industry), Internship / fellowship, PhD, Informal mentorshipYes
52
30/06/2026 22:43:57Patrick Kahardiprajapatrick.khdipraja@gmail.comPhD Student at Fraunhofer Heinrich Hertz InstituteYes - all of itExcited!Internship / fellowship, Informal mentorshipYeslinguistic interpretability, function vectors, currently also interested in training dynamicsThe Atlas of In-Context Learning: How Attention Heads Shape In-Context Retrieval Augmentation (https://arxiv.org/abs/2505.15807); Fast & Faithful Function Vectors (https://arxiv.org/abs/2606.05079)
53
30/06/2026 22:47:41Rahul Vashishtrahulvashisht290@gmail.comPhD student at IIT Madras, IndiaYes - workshops onlyExcited!Job (industry), Postdoc, Informal mentorshipYesTraining dynamics of neural networks, Mechanistic understanding of deep learning models from optimization perspective, https://openreview.net/pdf?id=yESGs2JgSu

https://arxiv.org/pdf/2307.13421
54
30/06/2026 22:48:15Miranda Muqing Miaomiaom@seas.upenn.eduPhD Student at the University of PennsylvaniaYes - all of itNeutralJob (industry), Internship / fellowship, Informal mentorshipYesMechanistic Interpretability
55
30/06/2026 22:58:37Joe Kwonjoekwon333@gmail.comIndependent researcherYes - all of itExcited!Maybehttps://www.formationresearch.com/secret-loyalties-whitepaper.pdf
56
30/06/2026 23:00:13Ruixuan Dengrdeng62@gatech.eduPhD Student at Georgia Institute of TechnologyYes - all of itExcited!Internship / fellowshipYesFeature Geometry; Manifold Hypothesis; Model Biology
57
30/06/2026 23:03:52Yan Zhouterryzhou@fas.harvard.eduMasters Student at HarvardYes - workshops onlyExcited!Job (industry), Internship / fellowship, PhD, Informal mentorshipMaybeMech interp, AI safety
58
30/06/2026 23:07:14Arnesh Batraarnesh23129@iiitd.ac.inUndergraduate at IIITDYes - all of itNeutralJob (industry), Internship / fellowship, PhD, Informal mentorshipYesMechanistic interpretability, representation learning, ssl. Ai safety https://arxiv.org/abs/2605.23033
59
30/06/2026 23:07:23Benjamin Etheridgebe <guess the symbol> robots.ox.ac.ukPhD candidate at the University of OxfordYes - all of itExcited!Job (industry), Internship / fellowship, Postdoc, Informal mentorshipYesAI security. Mech interp as mechanism for attacks under strong threat models Where do reasoning models refuse - https://arxiv.org/abs/2507.03167v4
60
30/06/2026 23:12:18Ambroise Odonnatambroiseodonnattechnologie@gmail.comPhD student at Huawei and InriaYes - all of itExcited!Job (industry)YesMechanistic interpretability, pretraining, finetuning, self-improvementhttps://arxiv.org/pdf/2602.06883
61
30/06/2026 23:30:22Riccardo Cadeiriccardo.cadei@ist.ac.atPhD Student at ISTA NoExcited!Job (industry), Internship / fellowship, PostdocYesMech Int x CausalityFrom tokens to policy: https://arxiv.org/abs/2606.17010
62
30/06/2026 23:34:06Ramneet Kaurramneet.kaur@sri.comResearcher at Stanford Research Institute (SRI)Yes - all of itExcited!Job (industry)YesSafety, interpretability, alignment of LLM models
63
30/06/2026 23:44:13Dong-Kyum Kimkdkyum531@gmail.comPostdoc at Max PlankYes - all of itExcited!Job (industry)YesOut-of-context reasoning, Loss of plasticity, Model EditingICML 2026, AI Engram: In Search of Memory Traces in Artificial Intelligence (https://arxiv.org/abs/2606.14997)
ICLR 2026, Bilinear representation mitigates reversal curse and enables consistent model editing (https://openreview.net/forum?id=pdNaYcApbz)
64
30/06/2026 23:45:06Tristan Daytristan.julian.day@gmail.comResearch Fellow at CBAI (Cambridge Boston Alignment Initiative)NoExcited!Job (industry), Internship / fellowship, Informal mentorshipMaybeIntrospection, personas, model psychology, model welfare, advancement of science via AI, AI applications to neuroscience, computational neuroscienceMy WIP work on personas https://tristan-day-research.github.io/assets/CBAI%20Poster%20of%20Tristan%20Day.pdf. or my ICML paper on LLM metacognition https://tristan-day-research.github.io/assets/Fine-Tuning%20Enhances%20Latent%20Metacognitive%20Capability%20in%20Language%20Models.pdf. poster: https://tristan-day-research.github.io/assets/ICML%20Poster%20Metacognition%20is%20latent%20in%20base%20models%20and%20enhanced%20by%20post-training%20and%20fine-tuning.pdf
65
30/06/2026 23:47:21Tristan Daytristan.julian.day@gmail.comCBAI Research Fellow (Cambridge Boston Alignment Initiative)Yes - all of itExcited!Job (industry), Internship / fellowship, Informal mentorshipMaybeIntrospection, personas, model psychology, model welfare, advancement of science via AI, AI applications to neuroscience, computational neuroscienceMy WIP work on personas https://tristan-day-research.github.io/assets/CBAI%20Poster%20of%20Tristan%20Day.pdf. or my ICML paper on LLM metacognition https://tristan-day-research.github.io/assets/Fine-Tuning%20Enhances%20Latent%20Metacognitive%20Capability%20in%20Language%20Models.pdf. poster: https://tristan-day-research.github.io/assets/ICML%20Poster%20Metacognition%20is%20latent%20in%20base%20models%20and%20enhanced%20by%20post-training%20and%20fine-tuning.pdf
66
30/06/2026 23:50:26Zekun Wangzekun@gatech.eduPhD Student at Georgia Institute of TechnologyNoExcited!Job (industry), Internship / fellowshipYesContinual learning, self-improving agents, compositionality, reinforcement learninghttps://arxiv.org/abs/2605.07078, https://arxiv.org/abs/2605.07076
67
30/06/2026 23:52:00Dong-Kyum Kimkdkyum531@gmail.comPostdoc at the Max Planck Institute for Security and PrivacyYes - all of itExcited!Job (industry)Yesout-of-context reasoning, reversal curse, model editing, loss of plasticityICML 2026, AI Engram: In Search of Memory Traces in Artificial Intelligence (https://arxiv.org/abs/2606.14997)
ICLR 2026, Bilinear representation mitigates reversal curse and enables consistent model editing (https://openreview.net/forum?id=pdNaYcApbz)
68
30/06/2026 23:52:06Ian Rios-Sialeriansebas@umich.eduIndependent ResearcherYes - workshops onlyExcited!Job (industry), Internship / fellowshipInformal menteesYesLLM Bias, Functional Concepts, Mech Interphttps://arxiv.org/abs/2601.06116
69
01/07/2026 00:15:57Chih-Hao Howard Hsub11902080@csie.ntu.edu.twUndergraduate student at National Taiwan UniversityYes - workshops onlyExcited!Internship / fellowship, PhD, Informal mentorshipYesMechanistic interpretability, representational geometry, causal interventions, model psychology, NeuroAITracing Psychometric Inference in Large Language Models: https://openreview.net/forum?id=v0OOhHuDNL
70
01/07/2026 00:16:29Shivanand Kundargishivanandkundargi992@gmail.comPhD Student at the University of Maryland, Baltimore County NoExcited!Internship / fellowship, Informal mentorshipYesActionable and mechanistic interpretability for Continual learning/generalization of frontier models and agentic systems. 1) Sequentially Acquiring Concept Knowledge to Guide Continual Learning (first author, accepted at CoLLAs 2026)

2) On the Role of Mechanistic Interpretability for Vision Language Prompt Learning (first author, accepted at ICML'26 Mech interp workshop)
71
01/07/2026 00:19:48Junaid Hasanjunaidhasan01@gmail.comPhD Student at the University of Washington, SeattleYes - workshops onlyExcited!Job (industry)YesMech Interp, Representation Theory, AI architectures, Graph Theory, Formalizationhttps://openreview.net/forum?id=26C2wzldtJ
72
01/07/2026 00:20:20Vedant Srinivasvedants8@stanford.eduUndergrad Student at StanfordYes - all of itExcited!Internship / fellowship, Informal mentorshipYesMechanistic Interpretability, Agentic optimization
73
01/07/2026 00:23:28Shervin Ghasemloushervinghasemlou@gmail.comAI Research ScientistYes - all of itNeutralJob (industry)MaybeTheoretical foundations of Mechanistic Interpretability, Biology inspired methodshttps://openreview.net/pdf?id=aRv6fhUJ2a
74
01/07/2026 00:33:33Changqing Fuchangqingfu@changqingfu.comIndependent ResearcherNoExcited!Job (industry), Informal mentorshipYesPrincipled Mech Interp for efficient architecture and learningchangqingfu.com/pdf/transformer.pdf
75
01/07/2026 00:42:37Cagatay Yildizcagatay.yildiz@uni-tuebingen.dePostdocYes - all of itExcited!Job (industry), Informal mentorshipYesunderstanding how continued pretraining changes the model internals and how models acquire knowledge via CPT
76
01/07/2026 01:21:00Melissa Wesselmelissawessel00@gmail.comIndependent researcherYes - workshops onlyExcited!Job (industry), Internship / fellowship, Informal mentorshipYesabstract/relational reasoning, variable binding, representational geometryIn-Context Learning Amplifies a Latent Symbolic Circuit -- https://openreview.net/attachment?id=9MKZdNBMC1&name=pdf
77
01/07/2026 01:41:25Sungjun Limlsj9862@gmail.comPhD student at Yonsei universityYes - all of itExcited!Internship / fellowshipYesTruthworthy AI; Reasoning; Mechanistic Interpretability; Variational Inference e.t.c.
78
01/07/2026 01:55:52Xi Chentonyxichen@gmail.comIncoming PhD Student at the Chinese University of Hong Kong, ShenzhenYes - all of itNeutralInternship / fellowship, Informal mentorshipYesCircuit Analysis, Latent Reasoning, Steering, Autoformalization
79
01/07/2026 01:57:02Saleena Angeline Sartawitasaleena.sartawita@mail.utoronto.caIndependent ResearcherYes - all of itExcited!Internship / fellowship, Informal mentorshipYesMechanistic Interpretability, AI Safety, Artificial General Intelligencehttps://arxiv.org/abs/2507.21141 - Death by a Thousand Directions: Exploring the Geometry of Harmfulness in LLMs through Subconcept Probing
80
01/07/2026 02:01:12Lasse Jantschlassejantsch@knu.ac.krMaster Student at Kyungpook National UniversityYes - workshops onlyExcited!Job (industry), Internship / fellowship, PhD, Informal mentorshipMaybeConcept representation, circuit tracing
81
01/07/2026 02:09:28Tiago Pimenteltiago.pimentel@inf.ethz.chPostdoc at ETH ZurichNoExcited!Job (industry)Informal menteesMaybeCausal abstraction, mech interp, AI safetyhttps://arxiv.org/pdf/2507.08802
82
01/07/2026 02:21:18Linyu Liuliulinyu4@huawei.comindependent researcherYes - all of itNeutralMaybemechanistic interpretabilityhttps://arxiv.org/abs/2605.18022
83
01/07/2026 02:24:07Arya Datlaavd42@cornell.eduUndergraduate Student at Cornell UniversityNoExcited!Internship / fellowship, PhD, Informal mentorshipYesCausal Abstraction, Applied Mech Interp, SteeringPLOT: Progressive Localization via Optimal Transport in Neural Causal Abstraction (https://arxiv.org/abs/2605.06979)
84
01/07/2026 02:33:51Nelson Alexnelsontharappel@gmail.comIndependent researcherNoExcited!Job (industry), Internship / fellowship, PhD, Postdoc, Informal mentorshipMaybeNeural Networks, Mechanistic Interpretability, Continuous Learning
85
01/07/2026 02:39:12Ningyu Xunyxu22@m.fudan.edu.cnPhD Student at Fudan UniversityYes - all of itExcited!Job (industry), PostdocMaybeMechanistic interpretability; Representation learning; Concept learning; Abstraction and reasoningEmergent Structured Representations Support In-Context Inference in Large Language Models (https://openreview.net/forum?id=cEcZoXflcm)

Revealing emergent human-like conceptual representations from language prediction (https://doi.org/10.1073/pnas.2512514122)
86
01/07/2026 02:39:24Dahee Kwondaheekwon@kaist.ac.krPhD Student at KAISTYes - all of itExcited!Internship / fellowship, PostdocYesMy research interests span vision recognition and visual generative AI, with a particular focus on analyzing and interpreting models’ internal representations. Ultimately, I aim to use insights from these analyses to identify fundamental limitations in existing models and develop better AI systems.

At the Mechanistic Interpretability Workshop, I will present our work on identifying and evaluating polysemanticity. I am currently especially interested in circuit-level interpretability for vision-language models. At the main conference, I will present our work on diagnosing and mitigating an internal bottleneck that limits output diversity in text-to-image generative models.

Moving forward, I hope to extend these research directions through deeper mechanistic analyses and broader evaluations across different models and architectures.		 I am always happy to connect and discuss research with others working on related—or even entirely different—topics!
87
01/07/2026 03:03:27Abhijeet Dhakaneasd.2007mail@gmail.comResearch Scientist - University of TennesseeYes - workshops onlyExcited!Job (industry)Informal menteesYesAI4Science - particularly accelerated material discovery, theory-in-loop experiments, mechanistic interpretability for material science problems.https://abhijeetdhakane.me/cv
88
01/07/2026 03:09:29Atsushi Magataamagata3@sompo-japan.co.jpChief Engineer at Sompo Japan Insurance Inc.Yes - workshops onlyExcited!Job (industry)MaybeInterpretability, Agentic workflow, Evaluations
89
01/07/2026 03:12:03Hua XUhxu401@connect.hkust-gz.edu.cnUG at HKUST(GZ)Yes - all of itExcited!Internship / fellowship, PhD, Informal mentorshipYesMechInterp, Neural-symbolic (Probabilistic Circuits (HMM, HCLT, etc)), Discrete Diffusion, Comp Neuro (replaying and preplaying),
90
01/07/2026 03:18:11Jinsu Kimjkim.jinsukim@gmail.comindependent researcherYes - all of itExcited!PhD, Informal mentorshipYesManifold preservation for faithful interpretabilityhttps://arxiv.org/abs/2604.28119
91
01/07/2026 03:32:30Wangxuan Fanfanwx@cuhk.edu.cnPhD Student at the Chinese University of Hong Kong (Shenzhen)NoExcited!MaybeLearning theory in Tabular Foundation Model of In-context learning; LLM agent social simulation on ECON.
92
01/07/2026 03:40:39Gustavo Sandovalgustavo.sandoval@nyu.eduLecturer at NYUYes - workshops onlyExcited!Job (industry), Postdoc, Informal mentorshipYesI'm interested in mechanistic interpretability for AI security, organized around what is called a "knowing-doing gap": models often represent security-relevant information internally without it actually governing what they generate. I use linear probes, steering vectors, SAEs, and crosscoders to find where that gap lives and try to repair it at inference time instead of retraining. Lately I've been pulled toward text-diffusion code models, which feel almost untouched from an interp angle, along with agentic safety drift over long interactions, and refusal mechanisms. - How Transparent is DiffusionGemma? (Engels, et. al. https://arxiv.org/pdf/2606.20560) — text-diffusion transparency CoT monitorability;

- DLM-Scope: Mechanistic Interpretability of Diffusion Language Models via SAEs (Wang, et. al. https://arxiv.org/pdf/2602.05859) — first SAE framework for DLMs; diffusion-time interventions outperforming AR steering is a promising area.

- Surgical Repair of Insecure Code Generation in LLMs (https://arxiv.org/abs/2604.16697) — my own, for context on the "knowing-doing gap" angle.
93
01/07/2026 03:43:05Kunalktalreja6@gatech.eduUndergrad, Georgia TechYes - all of itExcited!Internship / fellowship, PhDYesMech Interp mathematical foundations, Scientific Applicationshttps://arxiv.org/abs/2602.12026, https://arxiv.org/abs/2508.18567
94
01/07/2026 03:54:02PARTHA PRATIM SAHAtechnical.partha@gmail.comindependent researchYes - all of itExcited!Job (industry), Internship / fellowship, PhDYesAI alignment, Mechanistic Interpretability, AI Safetyhttps://arxiv.org/abs/2606.01060
95
01/07/2026 03:56:51Christian Kennethkennethasikinnn@gmail.comIndependent ResearcherNoExcited!Internship / fellowship, PhDMaybeMech Interp for Knowledge Discovery in Life Sciences & Representation Learninghttps://doi.org/10.48550/arXiv.2606.23443
96
01/07/2026 04:16:32Zehao Jinlunamos.thu@gmail.comMS Student at Georgia Institute of TechnologyYes - workshops onlyExcited!Job (industry), Internship / fellowship, PhD, Informal mentorshipYesMechInterp, AI Agents, AI Alignmenthttps://arxiv.org/abs/2605.05892
https://arxiv.org/abs/2606.14040
97
01/07/2026 04:35:36Vamshi Krishna Bonagirivamshi.12.2003@gmail.comPhD Student at MBZUAIYes - all of itExcited!Internship / fellowship, Informal mentorshipYesGeometry of Representations, Human Simulations The Linear Representation Hypothesis :)
98
01/07/2026 04:43:28Yi Huhuyi2002@stu.pku.edu.cnPhD Student at Peking UniversityYes - workshops onlyExcited!Job (industry), Internship / fellowshipMaybeMech Interp, OOD generalizationTransformers learn factored representations (https://arxiv.org/abs/2602.02385); Case-Based or Rule-Based: How Do Transformers Do the Math? (https://arxiv.org/abs/2402.17709)
99
01/07/2026 05:08:09Yong-Min Shinyongmin.shin@lgresearch.aiResearch scientist at LG AI researchYes - all of itExcited!Informal mentorshipMaybeApplications of explainable AI (including mechanistic interpretability techniques) to various fields, including graph transformers and chemistry models
100
01/07/2026 05:24:54Jaeyoon Seotonyno193@gmail.comMS Student at Dongguk UniversityYes - all of itExcited!NoVision-Text Alignment, Explainable AI, Object-Centric Learning