Women in AI Research (WiAIR)

Why don't bigger LLMs look more like the human brain? In the brain's language network, today's large models explain only slightly more variance than GPT-2 XL - and the reason says a lot about what those brain regions actually do. Dr. Greta Tuckute (Research Fellow at the Kempner Institute for the Study of Natural and Artificial Intelligence, Harvard) joins Jekaterina Novikova on Women in AI Research to unpack what brain-LLM alignment does and does not tell us. Greta works where neuroscience, cognitive science and AI meet - and when asked to keep only one of those three labels, AI is the first one she drops. The conversation covers how brain alignment develops over training, how sparse autoencoders can turn the "you are just comparing one black box to another" critique into something testable, and what memory experiments reveal about how meaning is stored - from why "pineapple" sticks in memory and "light" does not, to which sentence embeddings predict what people remember. In this episode: Brain alignment tracks formal linguistic competence, not reasoning - and it emerges after not much more than a developmentally plausible ~100M tokens, not 300B Why better next-word prediction stops meaning "more brain-like" once a model has mastered language Sparse autoencoder features plus surprisal: for one frontal brain region, surprisal alone does almost as well as 32,000 SAE features, while some voxels are captured by just six features- Brains and LLMs share the main, high-variance features of language - not the idiosyncratic ones Why learning from BPE tokens puts brain-LLM comparisons on "pretty shaky ground", and what should come next A distinctive meaning makes words and sentences memorable - and SBERT predicts human sentence memory better than the other embedding models tested From running a photography business at 14 to a PhD at MIT, and why trying the other path first was worth it PAPERS DISCUSSED From Language to Cognition: How LLMs Outgrow the Human Language Network Interpreting Brain Responses to Language with Sparse Features from Language Models Driving and suppressing the human language network using large language models Intrinsically memorable words have unique associations with their meanings A distinctive meaning makes a sentence memorable MENTIONED IN THIS EPISODE Anna Ivanova on formal vs functional linguistic competence (WiAIR) GRETA TUCKUTE Website: http://www.tuckute.com Bluesky: https://bsky.app/profile/gretatuckute.bsky.social X: https://x.com/GretaTuckute Women in AI Research (WiAIR) is a podcast and YouTube channel where Jekaterina Novikova talks with women doing AI research about their work, the questions driving it, and the paths that brought them there. 🎧 Subscribe to stay updated on new episodes spotlighting brilliant women shaping the future of AI. Follow WiAIR at: ⁠⁠⁠LinkedIn⁠⁠⁠ ⁠⁠⁠Bluesky⁠⁠⁠ ⁠⁠⁠X (Twitter)⁠⁠⁠ ⁠⁠⁠⁠WiAIR website⁠

An image is not worth a thousand words - it's worth an indefinite number of them. So why do the metrics we use to evaluate AI-generated image descriptions still assume there's one correct answer? In this episode of Women in AI Research, I talk with Elisa Kreiss (Assistant Professor of Communication at UCLA, director of the Coalas Lab) about what happens when you actually test the metrics the field relies on, and why CLIPScore, one of the most widely used measures for scoring image descriptions, stops correlating with human judgment the moment you introduce context. We also get into why longer descriptions aren't necessarily more informative, what happens when you just ask a model to "be concise," and whether AI models trip over charts and graphs the same way humans do. Elisa's research sits at the intersection of linguistics, accessibility, and multimodal AI, and this conversation covers the full arc of her work, from the theoretical question of why humans never describe images the same way twice, to the practical question of what that means for building systems that actually work for blind and low-vision users.In this episode: Why "context matters" is more radical than it sounds for image description evaluation The hidden reason CLIPScore breaks down once context enters the picture Why length is a bad proxy for information density - and what to use instead What happens when you prompt a model to just "be concise" Why charts and photos need completely different evaluation approaches Whether AI models make the same mistakes as humans when reading data visualizations What NeurIPS's Top Reviewer Award taught her about writing a genuinely useful peer review Resources & Links: Context Matters for Image Descriptions for Accessibility: Challenges for Referenceless Evaluation Metrics When More Words Say Less: Decoupling Length and Specificity in Image Description Evaluation CHART-6: Human-Centered Evaluation of Data Visualization Understanding in Vision-Language Models 🎧 Subscribe to stay updated on new episodes spotlighting brilliant women shaping the future of AI. Follow WiAIR at: ⁠⁠LinkedIn⁠⁠ ⁠⁠Bluesky⁠⁠ ⁠⁠X (Twitter)⁠⁠ ⁠⁠⁠WiAIR website⁠

Only 19% of Americans say AI has actually improved their productivity - so why the gap between the hype and reality? In this episode of Women in AI Research, Dr. Malihe Alikhani (Northeastern University, Contextual AI Lab) unpacks the hidden failures in how we build and deploy AI: why sycophancy is really a collapse of alignment, why bigger models aren't better aligned, and why "thin" alignment breaks down in the real world. Key topics The impact of moving across different AI contexts on system design The role of language as performative and active in shaping reality Interactive inference and uncertainty in AI systems The importance of context in meaning and system design AI policy, transparency, and societal impact Sycophantic behavior in large language models Measuring AI alignment: thin vs. thick AI adoption across sectors and demographic groups The role of policy in AI development and safety Ethical considerations in AI research and deployment Resources & Links: Breaking the AI Mirror: Sycophancy, productivity, and the future of collaboration Hype and harm: Why we must ask harder questions about AI and its alignment with human values How are Americans using AI? Evidence from a nationwide survey Connect with Dr. Malihe Alikhani: https://x.com/malihealikhani 🎧 Subscribe to stay updated on new episodes spotlighting brilliant women shaping the future of AI. Follow WiAIR at: ⁠LinkedIn⁠ ⁠Bluesky⁠ ⁠X (Twitter)⁠ ⁠⁠WiAIR website⁠

Do large language models truly understand language—or are they sophisticated pattern matchers? In this conversation, Dr. Anna Ivanova (Asst. Prof. at Georgia Tech) explores one of the important questions in AI: the relationship between language, thought, and intelligence. Drawing from neuroscience, cognitive science, and AI research, Anna explains why language understanding is harder to define than most people realize, why reasoning and language are not the same thing, and what today's LLMs can and cannot tell us about human cognition. Key Topics: Do LLMs understand language or merely generate convincing text? The difference between formal and functional linguistic competence What LLMs can learn from language alone—and what they cannot Why human cognition and AI cognition may be fundamentally different Theory of mind, reasoning, and common misconceptions about AI capabilities How cognitive scientists evaluate the "thinking" abilities of LLMs What neuroscience can teach AI researchers about interpretability Why understanding AI requires studying both behavior and internal representations The future of multimodal models and AI cognition Resources & Links: What does it mean to understand language? Dissociating language and thought in large language models How to evaluate the cognitive abilities of LLMs How Do LLMs Use Their Depth? True Lens Connect with Dr. Anna Ivanova: https://bsky.app/profile/neuranna.bsky.social https://x.com/neuranna 🎧 Subscribe to stay updated on new episodes spotlighting brilliant women shaping the future of AI. Follow WiAIR at: LinkedIn Bluesky X (Twitter) ⁠WiAIR website⁠

What actually happens when AI systems fail in the real world? In this final part of our conversation with Saadia Gabriel (UCLA), we unpack one of the most urgent challenges in modern AI: why even the most advanced models remain vulnerable to manipulation - and what that means for safety, fairness, and society. From multi-turn jailbreaking attacks with near 100% success rates to misinformation shaping human beliefs, this conversation goes beyond surface-level concerns and dives into how harms actually emerge in deployed systems. We explore: Why current guardrails are not enough How realistic attack scenarios differ from academic benchmarks The connection between model vulnerabilities and societal harm What AI can (and cannot) do about misinformation and persuasion The open research problems that still don’t have solutions Resources & Links: Generative AI in the Era of 'Alternative Facts' ModelCitizens: Representing Community Voices in Online Safety Translation as a Scalable Proxy for Multilingual Evaluation Connect with Dr. Saadia Gabriel: https://x.com/GabrielSaadia https://bsky.app/profile/skgabrie.bsky.social