Introducing Falcon-H1-Arabic: Pushing the Boundaries of Arabic Language AI with Hybrid Architecture
More from the blog
How an MIT research project became a global programming language
With millions of users across the world, Julia has been used to conduct cutting-edge research and to design new drugs, jet engines, heat pumps, and more.
The Open ASR Leaderboard Adds Its First Global South Language
How mobility gives language models a deeper understanding of place
Algorithms & Theory
Introducing Gemini 3.7 Flash
Gemini 3.7 Flash is our most intelligent workhorse model yet for coding and agents.
Related papers
Constitutional Midtraining: Content Presence Drives Alignment Gains
Post-training alignment is often shallow, eroding under fine-tuning. It remains untested as to whether constitutional midtraining interventions can produce durable alignment when cleanly isolated from post-training. We build a 394M-token constitutional corpus from Anthropic's Constitution and apply constitutional midtraining at 120B scale, where principled, values-based content is inserted into midtraining. A 2x2 design (curriculum ordering x deliberative reasoning) was used to produce four constitutionally midtrained conditions, plus a control, which were evaluated on self-generated and established benchmarks including alignment under pressure, value conflict resolution, blackmail, and emergent misalignment. All models were evaluated across three stages: post-midtraining, post-SFT, and post-benign fine-tuning. Constitutionally midtrained models outperformed the control on alignment generalization and durability, notably on blackmail: SFT instilled a blackmail propensity in all models, but constitutional midtraining blunted it, with the advantage surviving benign fine-tuning (-17.5pp). This durability did not extend to settings that required active resistance to in-context pressure or conflict, where the advantage attenuates after SFT. The presence of constitutional content at midtraining also mattered more than its structure, and constitutional midtraining incurred no capability cost, on average, at any stage (MMLU, ARC-Easy, piqa, GSM8K). A modest amount of constitutional content at midtraining could therefore yield broad, persistent alignment gains, offering a cheap, complementary addition to SFT-centered pipelines. Code, data, and models are available.
AI and Bullshit
It is argued that both AI and bullshitters are untrustworthy informants, and for similar reasons, it is natural to describe AI’s informational outputs as bullshit, as it signals their distinctive kind of epistemic deficiencies, which they share with bullshit.
Understanding Multilingual Medical ASR Adaptation Through Layer-Wise Analysis
This paper investigates how multilingual medical adaptation reshapes the internal representations of Whisper models through layer-wise encoder analysis, and shows that English medical fine-tuning produces the dominant encoder shift, whereas multilingual continuation largely preserves the adapted representation space.
Speculative Probing: LLM Monitoring at Speculative-Decoding Cost
It is found that the speculative-decoding module in recent LLMs can be repurposed for efficient high-quality classification by appending a trained soft prompt at the end of the target sequence, which can repurpose the speculative-decoding module into a sequence classifier.