Baidu killed the MTP module everyone copied from DeepSeek
This is a AI post classified by AGK Intelligence as Frontier models (a model release), kept by the AI Radar because it carries real work, not commentary.
Baidu killed the MTP module everyone copied from DeepSeek ERNIE 5.5 Flash just runs its own last layer a second time to guess the token after next. They call it ECHO. Trained on only 1 extra token and it still drafts several ahead for speculative decoding, w/ higher acceptance than an MTP module trained the same way With a normal MTP module the future token loss mostly trains the bolted on module. ECHO puts that loss on the real model so the backbone itself learns to look ahead, and Baidu says reasoning and code got clearly better. The leftover ECHO layer is like 10 times smaller than the MT
Posted by Binkg (14.3k followers) 1 h ago · 11 likes · 860 views · view the original post on X. Kept by the AI Radar as Frontier models.
More AI work like this
- Baidu just dropped ERNIE-5.5-Flash-Base, and the weights are available. — @kimmonismus
- Anthropic内部还有3个能力强于Opus 5.5的模型。根据Metr的披露，最强模型是Model 4，178.6分，而Opus 5.5 只有174.56分 — @mylifcc
- Voici un exemple de raisonnement stratégique que GPT-6 Astra met en œuvre en jouant… — @Fabien_Mikol
- Qwen 4.0要来了！ — @NFT_Chen
- Baidu killed the MTP module everyone copied from DeepSeek — @dummerspast39
- Only 6 of ERNIE 5.5 Flash Base's 43 layers see every token individually — @dummerspast39
- ALIBABA 🔥: "The becoming begins" teaser has been posted on the Qwen X account. We… — @testingcatalog
- Qwen 4??? — @niklaslenz_ai
Every post is read and classified by AGK Intelligence: what it is, which market it belongs to, and whether the link is a real tool. 65.7k posts from 5k X accounts over the last 14 days, 10.4k tools, 19 markets. New posts are collected several times an hour and the full ranking is recomputed every hour — last update 2026-11-06 13:30 UTC. Full method.