Meta tags:
author=Ali Kuwajerwala;
description=Ali s website, with links and everything. ;
keywords=ali, kuwajerwala, robotics, computer, science, artificial, intelligence, engineer, academic-website, portfolio-website;

Headings (most frequently used words):

ali, kuwajerwala, news, selected, publications,

Text of the page (most frequently used words):
and (60), the (19), for (12), open (10), set (9), robotics (8), #kuwajerwala (7), conceptfusion (7), with (6), planning (6), that (6), concepts (6), models (6), conceptgraphs (6), ali (5), maps (5), approaches (5), about (5), was (5), multimodal (4), krishna (4), alihusein (4), qiao (4), abs (4), scene (4), semantic (4), representation (4), language (4), scale (4), this (4), world (4), also (4), worked (4), 2023 (3), mapping (3), computer (3), jatavallabhula (3), murthy (3), paull (3), liam (3), shkurti (3), florian (3), arxiv (3), 2302 (3), 07241 (3), navigation (3), they (3), can (3), pre (3), these (3), using (3), work (3), reasoning (3), working (3), foundation (3), spatial (3), supervised (3), more (3), 2024 (3), icra (3), vocabulary (3), perception (3), robots (3), representations (3), which (3), are (3), apr (3), accepted (3), feel (3), free (3), out (3), roles (3), copyright (2), 2026 (2), international (2), year (2), rss (2), publisher (2), title (2), vision (2), intelligence (2), fos (2), information (2), sciences (2), tenenbaum (2), joshua (2), melo (2), celso (2), miguel (2), torralba (2), antonio (2), author (2), https (2), org (2), url (2), doi (2), misc (2), building (2), environment (2), closed (2), only (2), reason (2), training (2), time (2), recent (2), prompts (2), enabling (2), from (2), images (2), audio (2), demonstrate (2), features (2), via (2), multi (2), view (2), not (2), than (2), number (2), real (2), datasets (2), environments (2), task (2), website (2), pdf (2), bib (2), authors (2), morin (2), sacha (2), graphs (2), tasks (2), require (2), large (2), downstream (2), through (2), research (2), 2022 (2), reach (2), email (2), linkedin (2), you (2), like (2), philosophy (2), software (2), please (2), take (2), look (2), engineering (2), university (2), papers (2), there (2), powered, theme, last, updated, june, folio, jekyll, creative, commons, attribution, share, alike, science, systems, pattern, recognition, artificial, keywords, omama, mohd, chen, tao, shuang, iyer, ganesh, saryazdi, soroush, keetha, nikhil, tewari, ayush, madhava, 48550, conceptfusion2023, central, robot, interaction, objects, most, existing, integrate, largely, remain, confined, setting, finite, defined, further, queried, class, labels, text, address, both, issues, fundamentally, beyond, inherently, diverse, range, possible, queries, map, geometry, all, concert, leverages, capabilities, today, trained, internet, data, across, modalities, such, natural, pixel, aligned, fused, into, traditional, slam, fusion, enables, effective, zero, shot, needing, any, additional, finetuning, retains, long, tailed, better, outperforming, them, margin, iou, extensively, evaluate, simulated, home, tabletop, manipulation, autonomous, driving, platform, showcase, new, avenues, blending, conference, automation, sen, bipasha, agarwal, aditya, rivera, corban, paul, william, ellis, kirsty, chellappa, rama, gan, chuang, gu2023conceptgraphs, perform, wide, variety, semantically, rich, yet, compact, efficient, driven, have, attempted, leverage, encode, semantics, however, tend, produce, per, point, feature, vectors, well, larger, nor, contain, relationships, between, entities, useful, propose, graph, structured, scenes, built, leveraging, fusing, their, output, association, resulting, generalize, novel, classes, without, need, collect, finetune, utility, specified, abstract, complex, over, selected, publications, founded, head, operations, utm, club, 2019, received, nserc, undergraduate, student, award, mar, 2020, equivariant, paper, feb, preorders, launched, bot, jul, 2025, news, chat, physics, etc, code, stuff, fun, big, fan, karl, popper, david, deutsch, github, page, interned, where, llms, query, databases, chatgpt, during, undergrad, liquid, analytics, epson, visual, mobile, toronto, amazon, previously, deep, learning, humanoid, startup, before, got, master, small, steps, towards, cool, montréal, mila, labs, currently, actively, seeking, full, relocation, non, what, matters, important, problems, resume, don, hesitate, hqk, gmail, com, twitter, here, name, passionate, could, say, engaged, quest, understand, simulate, researcher, engineer, ctrl, podcast, projects, blog, current, toggle,

Text of the page (random words):
ali kuwajerwala toggle navigation about current blog cv papers projects podcast ctrl k ali kuwajerwala ai and robotics researcher and engineer hi there my name is ali and i m passionate about ai robotics and philosophy you could say i m engaged in the quest to understand and simulate intelligence i am currently actively seeking out full time roles i m open to relocation and non robotics roles what matters to me is working on important problems take a look at my resume here please don t hesitate to email me at ali hqk x x gmail com or on twitter linkedin previously i was working on deep learning research and engineering for humanoid robots at the startup k scale labs before that i got my master s at mila the university of montréal supervised by liam paull the conceptfusion and conceptgraphs papers which i worked on there are small steps towards building cool world models for robotics i also interned at amazon where i worked on using llms to query databases this was pre chatgpt during my undergrad at the university of toronto i worked on visual navigation for mobile robots supervised by florian shkurti i ve also worked in software engineering roles at epson and liquid analytics i also like to code stuff for fun feel free to take a look at my github page i m also a big fan of the work of karl popper and david deutsch if you d like to chat about philosophy ai software physics etc please feel free to reach out ️ feel free to reach out via email x or linkedin news jul 01 2025 k scale launched the k bot for preorders feb 01 2024 conceptgraphs was accepted to icra apr 05 2023 conceptfusion was accepted to rss ️ apr 16 2022 the equivariant representations paper was accepted to icra 2022 mar 31 2020 received the nserc undergraduate student research award apr 06 2019 co founded the utm robotics club as head of operations selected publications conceptgraphs open vocabulary 3d scene graphs for perception and planning alihusein kuwajerwala qiao gu sacha morin and 13 more authors 2024 abs bib pdf website for robots to perform a wide variety of tasks they require a 3d representation of the world that is semantically rich yet compact and efficient for task driven perception and planning recent approaches have attempted to leverage features from large vision language models to encode semantics in 3d representations however these approaches tend to produce maps with per point feature vectors which do not scale well in larger environments nor do they contain semantic spatial relationships between entities in the environment which are useful for downstream planning in this work we propose conceptgraphs an open vocabulary graph structured representation for 3d scenes conceptgraphs is built by leveraging 2d foundation models and fusing their output to 3d by multi view association the resulting representations generalize to novel semantic classes without the need to collect large 3d datasets or finetune models we demonstrate the utility of this representation through a number of downstream planning tasks that are specified through abstract language prompts and require complex reasoning over spatial and semantic concepts misc gu2023conceptgraphs title conceptgraphs open vocabulary 3d scene graphs for perception and planning author kuwajerwala alihusein and gu qiao and morin sacha and jatavallabhula krishna murthy and sen bipasha and agarwal aditya and rivera corban and paul william and ellis kirsty and chellappa rama and gan chuang and de melo celso miguel and tenenbaum joshua b and torralba antonio and shkurti florian and paull liam publisher international conference on robotics and automation icra url https arxiv org abs 2302 07241 year 2024 conceptfusion open set multimodal 3d mapping krishna murthy jatavallabhula alihusein kuwajerwala qiao gu and 13 more authors 2023 abs doi bib pdf website building 3d maps of the environment is central to robot navigation planning and interaction with objects in a scene most existing approaches that integrate semantic concepts with 3d maps largely remain confined to the closed set setting they can only reason about a finite set of concepts pre defined at training time further these maps can only be queried using class labels or in recent work using text prompts we address both these issues with conceptfusion a scene representation that is 1 fundamentally open set enabling reasoning beyond a closed set of concepts and ii inherently multimodal enabling a diverse range of possible queries to the 3d map from language to images to audio to 3d geometry all working in concert conceptfusion leverages the open set capabilities of today s foundation models pre trained on internet scale data to reason about concepts across modalities such as natural language images and audio we demonstrate that pixel aligned open set features can be fused into 3d maps via traditional slam and multi view fusion approaches this enables effective zero shot spatial reasoning not needing any additional training or finetuning and retains long tailed concepts better than supervised approaches outperforming them by more than 40 margin on 3d iou we extensively evaluate conceptfusion on a number of real world datasets simulated home environments a real world tabletop manipulation task and an autonomous driving platform we showcase new avenues for blending foundation models with 3d open set multimodal mapping misc conceptfusion2023 doi 10 48550 arxiv 2302 07241 url https arxiv org abs 2302 07241 author jatavallabhula krishna murthy and kuwajerwala alihusein and gu qiao and omama mohd and chen tao and li shuang and iyer ganesh and saryazdi soroush and keetha nikhil and tewari ayush and tenenbaum joshua b and de melo celso miguel and krishna madhava and paull liam and shkurti florian and torralba antonio keywords computer vision and pattern recognition cs cv artificial intelligence cs ai robotics cs ro fos computer and information sciences fos computer and information sciences title conceptfusion open set multimodal 3d mapping publisher robotics science and systems rss year 2023 copyright creative commons attribution share alike 4 0 international copyright 2026 ali kuwajerwala powered by jekyll with al folio theme last updated june 09 2026