2026
Bottlenecked Transformers: Periodic KV Cache Consolidation for Generalised Reasoning
ICLR 2026poster
Transformer LLMs have been shown to exhibit strong reasoning ability that scales with inference-time compute, most prominently through token-space “thinking” (i.e., chains of thought). A growing line of work pushes this extra computation into the model’s latent space (adjacent to standard decoding)…