Qwen3 Technical Report · HackerPapers discussions

Qwen3 Technical Report — mentioned by ElectricalUnion

Tue, 11 Aug 2026 00:11:59 GMT

You can use any message you want, but the model was tested to react reasonably well to the specific token sequence of "\nConsidering the limited time by the user, I have to give the solution based on the thinking directly now.\n</think>.\n\n" (from a Alibaba paper, struggling to find it now) Edit: arXiv:2505.09388 Qwen3 Technical Report

Paper and discussion history. A mention is not an endorsement.

Qwen3 Technical Report — mentioned by neongreen

Sat, 25 Jul 2026 18:27:48 GMT

Sounds similar to Qwen3 but the exact opposite: "stopping thinking before a natural termination" https://arxiv.org/pdf/2505.09388 > Thinking Budget. An additional advantage of Thinking Mode Fusion is that, once the model learns to respond in both non-thinking and thinking modes, it naturally develops the ability to handle intermediate cases—generating responses based on incomplete thinking. This capability lays the foundation for implementing budget control over the model’s thinking process. Specifically, when the length of the model’s thinking reaches a user-defined threshold, we manually…

Paper and discussion history. A mention is not an endorsement.

Qwen3 Technical Report — mentioned by virgildotcodes

Thu, 21 May 2026 04:22:18 GMT

> Synthetic training data is carefully crafted by humans. Much, if not the majority of synthetic data is AI generated. Human experts then evaluate samples of the data, but nothing like the entire corpus which can be trillions of tokens of generated material. See here where Qwen team discusses synthesizing trillions of tokens for their pre training dataset - https://arxiv.org/html/2505.09388v1 > The rare geniuses of human history use a different magnitude and configuration of the same kind of human intelligence I agree. What I don’t see any strong evidence for is that this intelligence is…

Paper and discussion history. A mention is not an endorsement.

Qwen3 Technical Report — mentioned by magicalhippo

Fri, 20 Feb 2026 16:05:04 GMT

From what I've gathered, they've been mostly training limited. Better training methods and cleaner training data allows smaller models to rival or outperform larger models training with older methods and lower-quality training data. For example, the Qwen3 technical report[1] says that the Qwen3 models are architecturally very similar to Qwen2.5, with the main change being a tweak in the attention layers to stabilize training. And if you compare table 1 in Qwen3 paper with table 1 in Qwen 2.5 technical report[2], the layer count, attention configuration and such is very similar. Yet Qwen3 was…

Paper and discussion history. A mention is not an endorsement.