Ndea

Watch the full interview: youtu.be/UyWSvLgQMDE

AI researcher @pidgeyusedgust of @ProseMsft joins us on the pod to discuss his favorite paper, "Semantic Programming by Example with Pre-trained Models" - a neurosymbolic framework where Flash Fill meets GPT-3. Symbolic for structure (syntactic), LLMs for meaning (semantic).

New open role: Technical Staff - Search Guidance Accelerate science & innovation. Join our small, talent-dense, globally remote team. Apply your RL/DL search expertise to the most advanced program synthesis system. Also: $10k referral bonus Details: ndea.com/jobs/search-gu…

Watch the mini-recap episode: youtu.be/t_RgBe2CVms Subscribe + share your feedback!

In the past 8 episodes of the Abstract Synthesis podcast, we've covered grammar filtering, temporal synthesis, inductive logic programming, vision-language programs, and symbolic world models. This week, we step back to reflect on 3 emergent themes.

Watch the full episode: youtu.be/O4Hyy-9VimI

"...simply looking at a problem differently can greatly improve learning performance." New pod: Ndea researcher & ILP expert @CelineHocquette on "Relational Decomposition for Program Synthesis" - making symbolic learning systems more efficient without domain-specific knowledge.

Watch the full episode: youtu.be/fCaLTOCiOsk

On the pod: @topwasu from @ellisk_kellis' lab at @Cornell discusses his paper PoE-World. We explore how symbolic world models can achieve strong generalization and sample efficiency by composing many small causal programs instead of learning a single monolithic model.

Dive into the paper, "Synthesizing Visual Concepts as Vision-Language Programs". A neuro-symbolic approach to visual concept induction that treats VLMs as perceptual tools inside symbolic programs. Watch the full episode: youtu.be/uefqvsButp8

Vision-language models (VLMs) can see well, but they struggle to reason. In this episode, @toniwuest (PhD researcher, @TUDarmstadt) explains how combining VLMs with program synthesis yields more reliable visual reasoning, with fewer tokens than chain-of-thought.

Watch the full interview: youtu.be/_8UJOYi-BKQ

Take a tour through the history & future of inductive logic programming (ILP) with Andrew Cropper. We discuss inductive bias, falsification-driven learning, solver-backed search, symbolic rule learning, and Popper, his influential modern ILP system.

Watch the full interview: youtu.be/Hw-PHdgFCEQ Or listen anywhere you get your podcasts.

On the pod: Professor @RiceUniversity and one of the most influential figures in logic, verification, and theoretical computer science, @vardi discusses his paper "Symbolic LTLf Synthesis" and the history of program synthesis.

Pre-show convo with the great @vardi about symbolic + neural reasoning for the Abstract Synthesis pod. Subscribe to watch the full interview next week. Don't miss this one.

Subscribe on YouTube or wherever you get your podcasts for upcoming episodes in 2026! youtu.be/yy_uqL86SSc

Special episode of the Abstract Synthesis podcast recorded live @ NeurIPS 2025. Program synthesis meets vision, quantum, RL, world models, and scientific discovery. Featuring @ClementBonnet16, @toniwuest, @LeopoldoSarra, @topwasu, Jumyung Park, Colin Conwell, Chris Hamblin.

Watch on YouTube: youtu.be/H5XDK6fJTtU Or anywhere else you get your podcasts.

New Abstract Synthesis podcast episode! Kedar Namjoshi of @BellLabs discusses his extended abstract "Program Synthesis And Non-monotonic Reasoning". Explore the gap between formal specifications and human expectations in IoT control programs + multi-agent robotics systems.