Skip to content
← All posts MIT projects selected for funding under US Department of Energy’s Genesis Mission

MIT projects selected for funding under US Department of Energy’s Genesis Mission

MIT News · Artificial Intelligence · news.mit.edu · By Office of the Vice President for Research · July 23, 2026

Initial research projects advance national priorities across natural resources, manufacturing, nuclear physics, and more.

Read on MIT News · Artificial Intelligence → Opens the original article in a new tab.

More from the blog

Related papers

#artificial intelligence Preprint Jul 2026

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.

Desiree Cho, Cameron Tice, Bernie Hogan et al. · 0 citations

Jailbreaking in the Haystack

NINJA (short for Needle-in-haystack jailbreak attack), a method that jailbreaks aligned LMs by appending benign, model-generated content to harmful user goals to reveal fundamental vulnerabilities in modern LMs.

R. Shah, C. Wu, Shashwat Saxena et al. · 4 citations

Convergent Evolution: How Different Language Models Learn Similar Number Representations

This paper identifies two different routes through which models can acquire geometrically separable features: they can learn them from complementary co-occurrence signals in general language data, including text-number co-occurrence and cross-number interaction, or from multi-token addition problems.

Deqing Fu, Tianyi Zhou, Mikhail Belkin et al. · 3 citations

SimulRAG: Simulator-based RAG for Grounding LLMs in Long-form Scientific QA

SimulRAG, a simulator-based RAG framework with a generalized retrieval interface that translates between text and simulator parameters/outputs, is proposed, which improves informativeness and factuality over the strongest adapted RAG baselines, while UE+SBA enhances claim-level efficiency and quality.

Haozhou Xu, D. Wu, M. Chinazzi et al. · 3 citations