Platonic Representation Hypothesis proven: 109M JEPA aligns with vision better than 72B LLMs
randall_balestr · x · 2026-10-30
Randall Balestrieri (with Yann LeCun) announces the Platonic Representation Hypothesis is now a theorem: separately trained JEPAs on any modalities provably learn the same representation up to one rotation. A 109M JEPA-Text model aligns with vision encoders better than LLMs 660x larger — scoring 0.184 with DINOv2 vs Qwen2.5-72B's 0.154, and 0.23 with CLIP's encoder vs the prior 0.21 SOTA. Alignment even follows a scaling law validated across 85 new runs.
More from Research
- Healthcare RL tasks are mostly billing code — where are the House-style diagnostic sims? — nirmal_dist · 2026-10-31
- Steering an "Einstein neuron" in Qwen 4B produces dramatically different answers — thursdai_pod · 2026-10-31
- Protein binder scoring head-to-head: ipDAE consistently strong, OpenDDE+ipTM surprisingly good at 80% PPV — AllThingsApx · 2026-10-31
- Turing TTS benchmark: best voice AI fools listeners only ~10% of the time — omarsar0 · 2026-10-31
- Ablation study: fine-grained MoE is the best LLM architecture change since GPT-3 — stochasticchasm · 2026-10-31
- Yann LeCun to Keynote MAIN2026 on World Models for Reasoning and Planning — karimjerbineuro · 2026-10-31