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 » The Unsolved Problem

The Unsolved Problem · October 2026

Self-improvement without labels: three papers try to build their own referees and curricula

Removing human labels means a system must pose its own problems, judge its own answers and decide what to learn next. Three new preprints each take on part of that job.

AI agentsMultimodal AI
A recursion needs two channels: the outer loop sets the inner loop’s objective, and the inner loop, Proposer–Director self-play on a randomized rollout group, returns the evidence for withdrawing skills.
A recursion needs two channels: the outer loop sets the inner loop’s objective, and the inner loop, Proposer–Director self-play on a randomized rollout group, returns the evidence for withdrawing skills. Figure from “R^2Orch: Recursive Self-Improvement for Agent Orchestration via Self-Play within a Skill-Internalization Loop” (arXiv)
RD
AI Research Digest research desk
October 2026 research · 5 min read · fact-checked against the source

Large models have improved mostly through outside signal: human labels, curated benchmarks and curricula designed by people. Getting them to keep improving without that help is still an open problem. A system that does it has to set its own practice problems, grade its own answers, and decide what to learn next. Three recent preprints work on parts of this. A-RSI swaps out tasks that have stopped teaching anything and checks answers with programs. R²Orch uses self-play to choose training tasks for agent orchestrators. EvoTeam changes how teams of agents are organized based on experience. This piece explains what is still missing and what a real solution would need to show.

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

Figures from the research

From truncation to completion. Both base decodes reach the limit for each item. finishes correctly in both decodes: 1,759/3,437 tokens for A1 and 5,140/2,515 for A2. Shorter, completed responses can improve benchmark accuracy even when the reasoning remains…
Figure 1. From truncation to completion. Both base decodes reach the limit for each item. finishes correctly in both decodes: 1,759/3,437 tokens for A1 and 5,140/2,515 for A2. Shorter, completed responses can improve benchmark accuracy even when the reasoning remains… Figure from “A-RSI: Adaptive Task Refresh for Zero-Data Multimodal Self-Improvement” (arXiv)
Three paradigms of MAS.
Figure 2. Three paradigms of MAS. Figure from “EvoTeam: Multi-Agent Collaboration via Evolving Teams” (arXiv)
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