( ˆ𐃷ˆ) .ᐟ.ᐟ
hi! i am a 3rd year PhD student in NLP @ MBZUAI advised by Monojit Choudhury, comparing how LLMs structure language to how humans do. i call it inductive linguistic behavioral interpretability of LLMs, or LLM linguistics on a casual day.
over the last two years, i've been cornering LLMs with controlled linguistic contrasts, watching how they behave, and working backwards to which distinctions they show, ignore, or get stumped by. so far that's been (1) what makes linguistics olympiad puzzles tricky for them (COLM, 2025), (2) whether they model human linguistic variation in Hindi-English verb code-mixing (EACL, 2026), (3) how they process hierarchical sequences compared to crows, monkeys, children, and adults (CogSci abstract, 2026; journal version cooking), (4) a taxonomy of translation difficulty for the Last Translation Benchmark, and (5) whether LMs make second-language learner-like mistakes across development (WIP). next up: a survey stitching these together, and neurodivergence x AI.
before this i completed my MSc. in Computational Linguistics at IIIT Hyderabad. linguistic complexity and how tech models it, was also the theme for my masters thesis on pseudowords generation (e.g. for LLM subword probes) and native speaker evaluation design in low-resource languages. this theme started inspiring me since i was a high schooler, solving actual linguistic reasoning puzzles for fun.
everything else i've worked on, is slotted below in the grid inspired by my thesis: contrasts i use to setup water-tight experiments, the behaviour i look out for and the interpretive tools i probe it with, and what i compare it against. rows run from language-first to mind-first, plus things that shipped in the wild. greens are done, purples are happening, blues are just starting off, join in! if a cell makes you go "huh?", that's on purpose, ask me :)
so, if you're into language, tech, comics, design, quizzing, board games, or just curious about what I'm up to, look around or hit me up! I keep it real, and it's always interesting :)
before this i completed my MSc. in Computational Linguistics at IIIT Hyderabad. linguistic complexity and how tech models it, was also the theme for my masters thesis on pseudowords generation (e.g. for LLM subword probes) and native speaker evaluation design in low-resource languages. this theme started inspiring me since i was a high schooler, solving actual linguistic reasoning puzzles for fun.
everything else i've worked on, is slotted below in the grid inspired by my thesis: contrasts i use to setup water-tight experiments, the behaviour i look out for and the interpretive tools i probe it with, and what i compare it against. rows run from language-first to mind-first, plus things that shipped in the wild. greens are done, purples are happening, blues are just starting off, join in! if a cell makes you go "huh?", that's on purpose, ask me :)
| contrastswhat i poke with | behavioral interpretabilitywhat i peek into, and pick apart | comparisonsagainst whom, across what | |
|---|---|---|---|
| language-first | puzzles in languages nobody trained ontranslations that fight back | code-mixed twitter morphologyfollowing the rules vs. spotting themtypology that showed up uninvitedwhat makes a translation hardfive ways a fact can go wrongtamil sounds, written down for STT | models × prompts × scalelanguage change, theirs and ourslanguage models from before they were large |
| in between | one verb slot, two languageswords that sound real but mean nothingmistakes: natural, learner-made, and fake | verb classes born from human disagreementhow real a fake word sounds to native earsskills hiding behind error patternshow surprised a model is | native speakers vs. model families, sizes, and mixescheckpoints, growing up like learners |
| mind-first | hierarchies vs. straight linessentences that shouldn't quite parsehow loneliness gets typed out | what people believe about their passwordswhat memory quietly does to number and casewhether MT helps the people who need it mostempathy, measured in entropymechanisms, mixed and weighed | crows, monkeys, kids, adults, and four LLM generationsneurodivergence × AIcreativity, theirs vs. ours |
| in the wild | synthetic data aimed at blind spotsprompts built to break things | chatbots that help the helpers, and the UI/UX they ship intickets, summarised and re-rankedexplanations for clusters that underperformguardrails for requirements that keep moving | fairness across who's speakingdialects across 9+ indian languages |
| across all of it | a survey of inductive linguistic behavioral interpretabilitydo the cells agree with each other? | ||
so, if you're into language, tech, comics, design, quizzing, board games, or just curious about what I'm up to, look around or hit me up! I keep it real, and it's always interesting :)
Action required
Problem: The current root path of this site is "baseurl ("_config.yml.
Solution: Please set the
baseurl in _config.yml to "