Skip to content
The AI Research Digest is included with Super Saver. See Super Saver
Tech AI Magazine Research
Digest
Sign in Subscribe

Research Digest » October 2026 research » Breakthrough 4 of 5

Breakthrough 4 of 5 · October 2026

Best AI agent solves 49 of 50 unpublished theoretical physics research problems

TheoPhysEval tests whether off-the-shelf agents can finish research-level physics calculations with no hints. The best one nearly sweeps it. The authors explain why that is not the same as doing research.

AI agents
Numerical solution of the differential equation.
Numerical solution of the differential equation. Figure from “TheoPhysEval: Assessing AI agents at autonomous theoretical physics problem solving” (arXiv)
RD
AI Research Digest research desk
October 2026 research · 3 min read · fact-checked against the source

TheoPhysEval is a set of 50 unpublished, research-level theoretical physics problems whose answers can be checked automatically. A world expert would typically need about a day or less for one. With web search and tools for scripting and symbolic work, the best agent, GPT-6-Astra in Codex, solved 49 of 50 at an average cost under USD 3 per problem. Opus 5.5 in Claude Code CLI solved 46. Removing internet access changed accuracy little for most agents, which the authors read as memorization of the literature. Each agent ran only once, and data leakage cannot be fully ruled out.

Super Saver

Keep reading, and get the Explain it Like I'm 12 version

The full article and its Explain it Like I'm 12 version are for Super Saver members. Sign in if you're a member.

Join Super Saver Sign in
More from October 2026

Keep exploring this issue

Figure S1: The Reed–Solomon decoder. (a) Measured forward Toffoli cost per step: one cycle of the 222-cycle synchronized EEA baseline of Khattar et al. [13] , or one round of our 70-round BM decoder. (b) Locator degree (top) and number of stored discrepancies…Paper of the Month

LLM-guided search finds cheaper quantum circuits in seven of eight benchmark tasks

A language model proposes circuit designs, a strict classical checker verifies and scores them, and the results beat prior designs while making a previously impractical classical algorithm usable in quantum form.

7 min+ ELI12
Positives that join less of a scene need more data ( Section 2.3 ). A synthetic world of five classes of six states (literal data, as in Figure 12 ): alignment of a text JEPA with the image side against the samples per state and the reach of its bad views…Breakthrough 1 of 5

A theorem for the Platonic Representation Hypothesis points to positive pairs, not scale

Balestriero and LeCun prove when separately trained JEPAs learn matching representations, then show a 109M-parameter text model aligning with vision better than a 72B language model.

Randall Balestriero, Yann LeCun4 min+ ELI12
Soft-MTP’s illustration, pseudocode, and loss performance compared to MTP. Blue line is the one-line code change . We refer readers to Algorithm 1 for the full code implementation of Soft-MTP, including the MTP loss weighting.Breakthrough 2 of 5

Soft-MTP: a one-line change strips noise out of multi-token prediction training

A new preprint finds that the standard way of training multi-token prediction heads hides a noisy gradient term. Removing it improved perplexity, speculative decoding and reasoning at no measurable extra cost.

Jayden Teoh4 min+ ELI12