IndexCache: Fixing Sparse Attention's Hidden O(L²) Problem

Shared entity: Drop / Same source domain / Shared topic / What happened next

ReBalance: Training-Free Efficient Reasoning for LRMs.

Both cover Drop; reported by the same outlet (arxiv.org); overlapping topics (accuracy, drop-in).

Shared entity: DeepSeek / Shared topic / What happened next

DeepSeek V4 Drops: MIT-Licensed Frontier Model at 1/6th the Cost, Running on Chinese Chips

Both cover DeepSeek; overlapping topics (attention, deepseek); picks up the DeepSeek thread on 2026-04-26.

DeepSeek V4 Preview Drops: 1.6T Parameters, 1M Context, Apache 2.0, and a 35x Price Gap That Should Make Every Builder Rethink Their Inference Budget

Both cover DeepSeek; overlapping topics (attention, deepseek); picks up the DeepSeek thread on 2026-04-24.

Shared entity: DeepSeek / Same source domain / What happened next

TERMINATOR: 14–55% CoT Token Reduction While Outperforming SOTA.

Both cover DeepSeek; reported by the same outlet (arxiv.org); picks up the DeepSeek thread on 2026-03-16.

Shared entity: DeepSeek / Same source domain / Earlier coverage

RAIM: Architecture-Aware Multi-Design Code Generation

Both cover DeepSeek; reported by the same outlet (arxiv.org); earlier DeepSeek coverage from 2026-03-03.

DualPath: DeepSeek Solves Agentic Inference Bottleneck (arXiv)

Both cover DeepSeek; reported by the same outlet (arxiv.org); earlier DeepSeek coverage from 2026-02-26.

Shared entity: DeepSeek / Shared topic / Earlier coverage

DeepSeek V4 Still Not Launched (Day 8+)

Both cover DeepSeek; overlapping topics (attention, deepseek); earlier DeepSeek coverage from 2026-02-25.

DeepSeek V4: Day 6, Still No Official Launch

Both cover DeepSeek; overlapping topics (attention, deepseek); earlier DeepSeek coverage from 2026-02-23.