Research
We trained language models that compress massive contexts into tiny latent representations. Latent Context Language Models (LCLMs) outperfor…
We trained language models that compress massive contexts into tiny latent representations. Latent Context Language Models (LCLMs) outperform existing KV cache compression methods on the latency/accur
We trained language models that compress massive contexts into tiny latent representations. Latent Context Language Models (LCLMs) outperform existing KV cache compression methods on the latency/accuracy frontier. 🧵1/10
Source: Yann LeCun (X) | 2026-06-09