Ziqiao "Martin" Ma

Here is a virtual greeting from Ziqiao. As many of my friends found it hard to pronounce my name (马子乔, pronounced as /ma˨˩˦ tsɨ˧˥ tɕʰiɑʊ˧˥/ in Mandarin), it's absolutely fine to just call me Martin alternatively.

CV (2025) / Google Scholar / Research / CSE595 NLP / GrowAI / Chat?

Bio / Profile Photo

Ziqiao (Martin) Ma is a 4th year Ph.D. candidate at the University of Michigan in Computer Science and Engineering advised by Professor Joyce Chai. His work has been supported in part by the Weinberg Cognitive Science Fellowship. He is also a part-time researcher at Adobe Research. Previously, he worked with Amazon Science. He received an Outstanding Paper Award at ACL 2023, and an Amazon Alexa Prize Award. He taught Natural Language Processing and won an Outstanding Graduate Student Instructor Award. He co-organized Bi-Align @ ICLR/CHI 2025, SpLU-RoboNLP @ ACL 2024, and co-instructed tutorial on Learning Language through Grounding @ NAACL 2025.

My Research (TL;DR)

Selected Awards/Recognitions

• Outstanding Paper Award, ACL 2023.
• First Place, Amazon Alexa Prize, 1st SimBot Challenge, 2023.
• Outstanding GSI Award, University of Michigan, 2022.
• James B. Angell Scholar, University of Michigan, 2021.

Selected Fellowships/Scholarships

• Weinberg Cognitive Science Fellowship, 2024.
• Rackham Doctoral Intern Fellowship, 2024.
• John Wu & Jane Sun Excellence Scholarship, 2017.

Updates

News

• [Pinned] We are founding GrowAI, an open-source community uniting researchers interested in human-like artificial general intelligence and growing AI like a child at scale.
• [Pinned] I will join ACL Year-Round Mentorship starting from 2025. Join our monthly mentoring sessions and let's grow together :)
• [Jan. 2025] The 1st Workshop on Bidirectional Human-AI Alignment (Bi-Align) will be with ICLR 2025, with a Special Interest Group session at CHI 2025. Look forward to your best work and join us in Singapore and Yokohama :)
• [Dec. 2024] The Tutorial on Learning Language through Grounding will be with NAACL 2025. See you in Albuquerque :)
• [Aug. 2024] I will be the Graduate Student Instructor for CSE 595 (NLP) in Fall 2024 at Michigan. Check out our LLM-edition of the course!
• [July 2024] Excited to be selected to receive the Weinberg Cognitive Science Fellowship!!
• [Jan. 2024] I started my intern with Adobe Research!!
• [Oct. 2023] The 4th International Combined Workshop on Spatial Language Understanding and Grounded Communication for Robotics (SpLU-RoboNLP) will be with ACL 2024. Look forward to your best work and join us in Bangkok :)
• [Jul. 2023] Our paper "World-to-Words: Grounded Open Vocabulary Acquisition through Fast Mapping in Vision-Language Models" was selected for the Outstanding Paper Award at ACL 2023!!
• [Jun. 2023] We won the 1st place in the Alexa Prize SimBot Challenge!!

Paper Alerts

• [Mar. 2025] One demo to appear in CVPR 2025, see you in Nashville :)
• [Jan. 2025] One paper to appear in NAACL 2025, see you in Albuquerque :)
• [Jan. 2025] One paper to appear in ICLR 2025 as an oral presentation, see you in Singapore :)

Seminar Talks

• [20250416] Grounding Lexical Semantics in the Era of Vision-Language Models @ Theoretical and Computational Neuroscience Journal Club, JHU.
• [20250326] Language Grounding to the Visual World and Human Interactions: How Far Are We from Embodied Dialogue Agents @ HAAG Seminar, Georgia Institute of Technology.
• [20250206] Bridging Minds and Machines: Cognitive Insights for Developing and Evaluating AI Systems @ Foreseer Group, UMich.
• [20241205] Language Grounding to the Visual World and Human Interactions: How Far Are We from Embodied Dialogue Agents @ University of Washington.

Experiences

Education

• Ph.D. in Computer Science and Engineering, Certificate in Cognitive Science
  University of Michigan, Ann Arbor | 2026 (Expected)
  Advisor: Joyce Chai
• BSE in Computer Science, Minor in Mathematics (Summa Cum Laude)
  University of Michigan, Ann Arbor | 2021
  

Industry

• MIT-IBM Watson AI Lab, Cambridge | May 2025 -
• Adobe Research, San Jose | Jan 2024 - May 2025
• Amazon Science, Sunnyvale | May 2023 - Aug 2023

Current Teaching

• [EECS 595 / SI 561 / LING 541), Natural Language Processing (Fall 2022, Fall 2024), University of Michigan.
  • Links: [Slides] [Homework] (To be published in Fall 2025, stay tuned)
• [EECS 492] Introduction to Artificial Intelligence (Winter 2021), University of Michigan.
  • Links: [Prolog Tutorial]

Guest Lectures

• [CSE 895], Selected Topics on Large Language Models (Mar. 2025), Michigan State University. Host: Parisa Kordjamshidi
• [DSC 250], Advanced Data Mining (Mar. 2025), University of California San Diego. Host: Zhiting Hu
• [COMPSCI 396], Reasoning and Planning in the Foundation Model Era (Feb. 2025), Northwestern University. Host: Manling Li
• [DSC 291], Machine Learning with Few Labels (May. 2024), University of California San Diego. Host: Zhiting Hu
• [PSYCH 745], Psychology of Language (Nov. 2023), University of Michigan. Host: Julie Boland
• [EECS 692], Advanced Artificial Intelligence (Jan. 2022; Feb. 2025), University of Michigan. Host: Joyce Chai

Academic Services

The 1st Workshop / Special Interest Group on Bidirectional Human-AI Alignment (Bi-Align @ ICLR 2025 / CHI 2025)

Co-organizer

[Homepage/CFP] [OpenReview]

The 4th International Combined Workshop on Spatial Language Understanding and Grounded Communication for Robotics (SpLU-RoboNLP @ ACL 2024)

Co-organizer

[Homepage/CFP] [OpenReview] [Proceedings]

The 5th Michigan AI Symposium: AI & Accessibility (2022)

Poster/Demo Chair

[Homepage/CFP]

• Mentor, ACL Year-Round Mentorship Program, since 2025.
• Seminar Tsar, Michigan AI Seminar Series, 2023-2025.
• Area Chair of ACL ARR (Multimodality and Language Grounding Track).
• Program Committee/Reviewer: NLP venues (e.g., ARR, ACL, EMNLP, NAACL, EACL, COLING, COLM, ...), ML/DM venues (e.g., ICLR, ICML, NeurIPS, AISTATS, KDD, ECML, TMLR, DMLR, TNNLS...), CV venues (ICCV, CVPR, ECCV, ...), Cognitive Science venues (COGSCI), and sometimes Robotics/HCI/CSS venues (RA-L, HRI, CHI, ICWSM, ...).

• Personal

Publications [.bib]

Show by... ( Recognition / Recent Selection / Cognitive AI Selection / Year / Topics )

Research Topics: Multimodal Learning and Generation / (Inter)active Learning and Alignment / Embodiment and Situated Intelligence / Teaching & Community Services

* indicates equal contributions; § indicates correspondence and mentoring.

Coming Soon

Coming Soon.

CVPR 2025 (Demo Track)

Paper / Homepage / GitHub / Dataset

TL;DR...

• Coming Soon.

Coming Soon

Coming Soon.

CCN 2025 (Extended Abstract)

Paper / Homepage / GitHub / Dataset

TL;DR...

• Coming Soon.

Coming Soon

Coming Soon.

ACL 2025 (Findings) / The 1st Workshop on World Models: Understanding, Modelling and Scaling @ ICLR 2025

Paper / Homepage / GitHub / Dataset

TL;DR...

• Coming Soon.

Training Turn-by-Turn Verifiers for Dialogue Tutoring Agents: The Curious Case of LLMs as Your Coding Tutors

Jian Wang, Yinpei Dai, Yichi Zhang, Ziqiao Ma, Wenjie Li, Joyce Chai

ACL 2025 (Findings)

Paper / GitHub

TL;DR...

• We introduce a novel agent workflow called Trace-and-Verify (TRAVER), which combines knowledge tracing to estimate a student's knowledge state and turn-by-turn verification to ensure effective guidance toward task completion.
• We introduce the coding tutoring task as the testbed for tutoring LLM agents.
• TRAVER effectively enables inference-time scaling for tutoring agents.

Vision-Language Models Are Not Pragmatically Competent in Referring Expression Generation

Ziqiao Ma*, Jing Ding*, Xuejun Zhang, Dezhi Luo, Jiahe Ding, Sihan Xu, Yuchen Huang, Run Peng, Joyce Chai

Preprint, 2025

Paper / Homepage / GitHub / Dataset

TL;DR...

• We introduce RefOI, a new dataset of 1.5k objects, each with 3 written and 2 spoken human-produced referring expressions.
• We also release RefOI-TLHF, a large dataset of token-level human feedback for 10.6k referring expressions.
• We identify three key failures of pragmatic competence in VLMs: (1) cannot uniquely refer to the referent, (2) include excessive or irrelevant information, and (3) misalign with human pragmatic preferences.

VEGGIE: Instructional Editing and Reasoning Video Concepts with Grounded Generation

Shoubin Yu*, Difan Liu*, Ziqiao Ma*, Yicong Hong, Yang Zhou, Hao Tan, Joyce Chai, Mohit Bansal

Preprint, 2025

Paper / Homepage / Dataset

TL;DR...

• We introduce VEGGIE, a diffusion-loss only video generative model that handles various tasks for both video concept grounding and editing from user instructions.
• Pixel-level grounded training helps various video concept editing task in multi-task learning.
• VEGGIE shows emergent zero-shot multimodal instructional and in-context video editing.

Do Vision-Language Models Represent Space and How? Evaluating Spatial Frame of Reference under Ambiguities

Zheyuan Zhang*, Fengyuan Hu*, Jayjun Lee*, Freda Shi, Parisa Kordjamshidi, Joyce Chai, Ziqiao Ma§

ICLR 2025 (Oral) / The 1st Pluralistic Alignment Workshop @ NeurIPS 2024

Paper / Homepage / GitHub / Dataset / Poster

TL;DR...

• We introduce COMFORT, a protocol to evaluate spatial reasoning in VLMs across multilingual and ambiguous frames of reference (FoR);
• VLMs exhibit poor robustness and consistency, lack the flexibility to accommodate multiple FoRs, and fail to adhere to language-specific or culture-specific conventions in cross-lingual tests.

Babysit A Language Model From Scratch: Interactive Language Learning by Trials and Demonstrations

Ziqiao Ma*, Zekun Wang*, Joyce Chai

NAACL 2025 / The 1st Workshop on LLMs and Cognition (LLMCog) @ ICML 2024 (Oral)

Paper / GitHub / Poster

TL;DR...

• We introduce a trial-and-demonstration (TnD) learning framework that incorporates three components: student trials, teacher demonstrations, and a reward conditioned on language competence at various developmental stages;
• TnD accelerates word representation learning for student models of equal and smaller numbers of parameters, and both trials and demonstrations matter.
• We further show that the teacher's choices of words influence students' word-specific learning efficiency, and a practice-makes-perfect effect is evident by a strong correlation between the frequency of words in trials and their respective learning curves.

Humanity's Last Exam

Community Contribution, lead by Center for AI Safety and Scale AI.

Technical Report, 2025

Paper / Homepage / Dataset

TL;DR...

• A multi-modal benchmark at the frontier of human knowledge, designed to be the final closed-ended academic benchmark of its kind with broad subject coverage;
• State-of-the-art LLMs demonstrate low accuracy and calibration, gaps remain compared to the expert human frontier on closed-ended academic questions.

Towards Bidirectional Human-AI Alignment: A Systematic Review for Clarifications, Framework, and Future Directions

Hua Shen, Tiffany Knearem, Reshmi Ghosh, Kenan Alkiek, Kundan Krishna, Yachuan Liu, Ziqiao Ma, Savvas Petridis, Yi-Hao Peng, Li Qiwei, Sushrita Rakshit, Chenglei Si, Yutong Xie, Jeffrey P. Bigham, Frank Bentley, Joyce Chai, Zachary Lipton, Qiaozhu Mei, Rada Mihalcea, Michael Terry, Diyi Yang, Meredith Ringel Morris, Paul Resnick, David Jurgens

Preprint 2024

Paper

TL;DR...

• We present a conceptual framework of Bidirectional Human-AI Alignment to organize the literature from a human-centered perspective;
• We survey studies of aligning AI to humans, that ensures AI produces the intended outcomes determined by humans;
• We survey studies of aligning humans to AI, which aims to help individuals and society adjust to AI advancements both cognitively and behaviorally.

Vision-and-Language Navigation Today and Tomorrow: A Survey in the Era of Foundation Models

Yue Zhang*, Ziqiao Ma*, Jialu Li*, Yanyuan Qiao*, Zun Wang*, Joyce Chai, Qi Wu, Mohit Bansal, Parisa Kordjamshidi

TMLR 2024 (Survey Certificate)

Paper / GitHub

TL;DR...

• We provide a top-down review on using foundation models to address VLN challenges;
• The roles of foundation models are divided into world model, human model, and the VLN agent.

Multi-Object Hallucination in Vision-Language Models

Xuweiyi Chen*, Ziqiao Ma*, Xuejun Zhang*, Sihan Xu, Shengyi Qian, Jianing Yang, David Fouhey, Joyce Chai

NeurIPS 2024 / The 3rd Workshop on Advances in Language and Vision Research (ALVR) @ ACL 2024

Paper / Homepage / GitHub / Dataset / Poster

TL;DR...

• We introduce ROPE, an evaluation protocol for hallucination across multiple objects using visual referring prompts;
• VLMs hallucinate more with multiple objects, are influenced by object class distribution, and exhibit behavior driven by data-specific and intrinsic factors.

DriVLMe: Enhancing LLM-based Autonomous Driving Agents with Embodied and Social Experiences

Yidong Huang, Jacob Sansom, Ziqiao Ma§, Felix Gervits, Joyce Chai

IROS 2024 / The 1st Vision and Language for Autonomous Driving and Robotics Workshop (VLADR) @ CVPR 2024

Paper / Hompage / Model / Poster

TL;DR...

• We introduce DriVLMe, a video-language-model-based agent to facilitate natural and effective communication between humans and autonomous vehicles that perceive the environment and navigate;
• We develop DriVLMe from both embodied experiences in a simulated environment and social experiences from real human dialogue.

GroundHog: Grounding Large Language Models to Holistic Segmentation

Yichi Zhang, Ziqiao Ma, Xiaofeng Gao, Suhaila Shakiah, Qiaozi Gao, Joyce Chai

CVPR 2024

Paper / Homepage / Model (Coming) / Dataset / Poster

TL;DR...

• We introduce GroundHog, a multimodal large language model grounded in holistic segmentation, using a masked feature extractor and unified grounding masks for fine-grained visual understanding.
• Trained on the curated M3G2 dataset, GroundHog outperforms in language grounding tasks, reduces object hallucination, and offers improved diagnosis for complex visual inputs.

Inversion-Free Image Editing with Language-Guided Diffusion Models

Sihan Xu*, Yidong Huang*, Jiayi Pan, Ziqiao Ma§, Joyce Chai

CVPR 2024

Paper / Homepage / GitHub / Live Demo / Poster

TL;DR...

• We derive Denoising Diffusion Consistent Model (DDCM), showing that when the initial sample is known, a special variance schedule reduces the denoising step to the same form as the multi-step consistency sampling;
• DDCM implies a inversion-free strategy without explicit inversion in sampling for image editing;
• We further unify the attention control mechanisms in an inference time algorithm for text-guided editing, taking less than 3 seconds per edit.

CycleNet: Rethinking Cycle Consistency in Text-Guided Diffusion for Image Manipulation

Sihan Xu*, Ziqiao Ma*, Yidong Huang, Honglak Lee, Joyce Chai

NeurIPS 2023

Paper / Homepage / GitHub / Dataset / Poster

TL;DR...

• We introduce CycleNet from theoretic derivations, a model that incorporates cycle consistency (and a self-supervision loss) into diffusion model to regularize image manipulation;
• CycleNet is robust even with very limited training data (around 2k) and requires minimal computational resources (1 GPU) to train.

Towards A Holistic Landscape of Situated Theory of Mind in Large Language Models

Ziqiao Ma, Jacob Sansom, Run Peng, Joyce Chai

EMNLP 2023 (Findings)

Paper / GitHub / Dataset / Poster

TL;DR...

• We taxonomize machine ToM into 7 mental state categories and delineate existing benchmarks to identify under-explored aspects of ToM;
• Pilot studies for a holistic and situated evaluation of ToM to break ToM into individual components and treat LLMs as an agent who is physically situated in environments and socially situated in interactions with humans.

Towards Collaborative Plan Acquisition through Theory of Mind Modeling in Situated Dialogue

Cristian-Paul Bara, Ziqiao Ma, Yingzhuo Yu, Julie Shah, Joyce Chai

IJCAI 2023

Paper / GitHub / Dataset / Poster

TL;DR...

• We study collaborative plan acquisition in human-AI tasks, where agents predict missing task knowledge for themselves and their partners using perceptual and dialogue history;
• We show that predicting a partner's missing knowledge, coupled with explicit modeling of dialogue moves and mental states, leads to better collaboration.

World-to-Words: Grounded Open Vocabulary Acquisition through Fast Mapping in Vision-Language Models

Ziqiao Ma*, Jiayi Pan*, Joyce Chai

ACL 2023 (🏆 Outstanding Paper Award)

Paper / GitHub / Model / Dataset / Poster

TL;DR...

• We introduce OctoBERT, a visually grounded language model designed to acquire grounding ability during pre-training and enable fast mapping of new words through few-shot learning without explicit grounding supervision;
• Visual grounding accelerates grounded word representation learning;
• Imageability aligns positively with human intuition and prediction metrics, while concreteness shows opposite correlations -> need for language learning agents to acquire word meanings through physical interactions!

NLP Reproducibility For All: Understanding Experiences of Beginners

Shane Storks, Keunwoo Peter Yu, Ziqiao Ma, Joyce Chai

ACL 2023 (Theme Track)

Paper / GitHub / Poster

TL;DR...

• We studied 93 NLP students replicating recent NLP papers;
• Programming skills and paper comprehension had limited impact on effort, while accessible documentation, coding practices, and data availability were critical.

SEAGULL: An Embodied Agent for Instruction Following through Situated Dialog

Yichi Zhang, Jianing Yang, Keunwoo Peter Yu, Yinpei Dai, Shane Storks Yuwei Bao, Jiayi Pan, Nikhil Devraj, Ziqiao Ma, Joyce Chai

Alexa Prize SimBot Challenge 2023 (🏆 1st Place)

Paper

TL;DR...

• SEAGULL is the winning solution to the Alexa Prize SimBot Challenge;
• SEAGULL features a modular system combining neural and symbolic components for language understanding, vision, state tracking, and policy execution.

DOROTHIE: Spoken Dialogue for Handling Unexpected Situations in Interactive Autonomous Driving Agents

Ziqiao Ma, Benjamin VanDerPloeg*, Cristian-Paul Bara*, Huang Yidong*, Eui-In Kim, Felix Gervits, Matthew Marge, Joyce Chai

EMNLP 2022 (Findings)

Paper / GitHub / Dataset / Poster

TL;DR...

• We present DOROTHIE, a simulation platform for studying sensorimotor-grounded dialogue in dynamic autonomous driving scenarios;
• We create the Situated Dialogue Navigation (SDN), a continuous outdoor VLN benchmark with real human dialogue.

DANLI: Deliberative Agent for Following Natural Language Instructions

Yichi Zhang, Jianing Yang, Shane Storks, Jiayi Pan, Nikhil Devraj, Ziqiao Ma, Keunwoo Peter Yu, Yuwei Bao, Joyce Chai

EMNLP 2022 (Oral)

Paper / GitHub / Slides

TL;DR...

• We propose DANLI, a neuro-symbolic deliberative agent for embodied AI that proactively reasons and plans using neural and symbolic representations;
• DANLI outperforms reactive baselines by over 70% on the TEACh benchmark.

Partition-Based Active Learning for Graph Neural Networks

Jiaqi Ma*, Ziqiao Ma*, Joyce Chai, Qiaozhu Mei

TMLR 2023 / The 8th International Workshop on Deep Learning on Graphs (DLG) @ KDD 2022 (Oral)

Paper / GitHub / Poster

TL;DR...

• We introduce GraphPart, a partition-based active learning method for GNN that selects representative nodes from graph partitions for querying;
• GraphPart is motivated by classification error analysis under smoothness assumptions;
• GraphPart outperforms existing active learning methods across benchmarks and budget constraints and reduces the accuracy disparity compared to random training node selection across most datasets.

If you like my figures here, I highly recommend you also visit SiX's homepage.

Misc

Fun Facts

• I was given my Chinese name, 马子乔, through a visually symbolic process, breaking down the radicals of 骄子, which roughly translates to 'the gifted child' in English. This is one of the reasons why logographic languages are so beautiful.
• I was born and raised up in Chengdu, the home of pandas. I am proud of Chengdu Foreign Languages School, my high school, and identify myself as a CFLSer. 成外人永远不会成外人.
• I am INFJ according to the Myers-Briggs, and my friends said that I exhibit stereotypical traits of this personality type...lol
• I love literature and plays. I am particularly interested in Shakespeare's plays, traditional Chinese opera, Latin American literature, and modern Asian literature.
• I love movies, I am obsessed with the Czechoslovak New Wave and psychological thrillers these days.
• I am a super fan of Nightwish. My favourite album is Imaginaerum.
• [NEW] Lately, I've become deeply fascinated with herbariums. I just started to keep a personal herbarium journal!

Game Design (More)

I seriously considered a career in game design, and although I ultimately chose a different path, it provided excellent preparation for my work in embodied AI research, which often involves intensive programming with simulators.

• a list of student-made games I enjoyed: Mogu, Turbo Neon.
• a list of video games I enjoyed: Sandbox (MineCraft, Terraria), Story-based RPG & Interactive Stories (Stardew Valley, Undertale, The Stanley Parable, This War of Mine, To the Moon, Season: A Letter to the Future), Roguelike (Soul Knight, Risk of Rain), Puzzle & Puzzle-based Platformer (Gorogoa, Limbo, Inside, Chants of Sennaar), Battle Royales (Naraka Bladepoint, Fall Guys).

Contracts

Zekai Fan, Shiyu Qu, Juan Rivera Plata, Yihao Huang, Ziqiao Martin Ma

[trailer][itch.io][indidb][tigsource]

• A turn-based tactic video game.

Floating

Ziqiao Martin Ma

[play online][download]

• A physics-based puzzle video game.

Mentoring

Random Tours

• Take a random virtual stroll over to one of my friends' homepage! It's like a digital house call, minus the awkward small talk and the "sorry, my place is a mess" excuse! Let's go!
• When I was exhausted but couldn't take time off to travel, I'd go on virtual adventures instead: randomly searching for remote destinations and dropping pins on Google Maps. Here are a few spots that I swear I'll visit in person...someday, eventually! どこでもドア
• Why I am still staying alive? Here's why

Chat?

If you would like to have a random (virtual) coffee chat with me, please visit my calendly page. I am happy to talk if you want to share your stress or just want to chat about life in general (when I have time), but be sure to check out the On-Campus Mental Health Resources @ Michigan.

Get In Touch

You are welcome to drop me a message :)