Agentic Discovery of Neural Architectures: AIRA-Compose and AIRA-Design

Agentic Discovery of Neural Architectures: AIRA-Compose and AIRA-Design | Signal Canvas | ScienceToStartup

Page Freshness

Signal Canvas proof surface

Canonical route: /signal-canvas/agentic-discovery-of-neural-architectures-aira-compose-and-aira-design

stale

Proof freshness: stale
Proof status: unverified
Display score: 9/10
Last proof check: 2026-05-18
Score updated: 2026-05-18
Score fresh until: 2026-06-17
References: 0
Source count: 4
Coverage: 50%

This page is showing the last landed evidence receipt and score bundle because the latest proof data is outside the freshness window.

Agent Handoff

Canonical ID agentic-discovery-of-neural-architectures-aira-compose-and-aira-design | Route /signal-canvas/agentic-discovery-of-neural-architectures-aira-compose-and-aira-design

REST example

curl https://sciencetostartup.com/api/v1/agent-handoff/signal-canvas/agentic-discovery-of-neural-architectures-aira-compose-and-aira-design

MCP example

{
  "tool": "search_signal_canvas",
  "arguments": {
    "mode": "paper",
    "paper_ref": "agentic-discovery-of-neural-architectures-aira-compose-and-aira-design",
    "query_text": "Summarize Agentic Discovery of Neural Architectures: AIRA-Compose and AIRA-Design"
  }
}

source_context

{
  "surface": "signal_canvas",
  "mode": "paper",
  "query": "Agentic Discovery of Neural Architectures: AIRA-Compose and AIRA-Design",
  "normalized_query": "2605.15871",
  "route": "/signal-canvas/agentic-discovery-of-neural-architectures-aira-compose-and-aira-design",
  "paper_ref": "agentic-discovery-of-neural-architectures-aira-compose-and-aira-design",
  "topic_slug": null,
  "benchmark_ref": null,
  "dataset_ref": null
}

Evidence Receipt

Route status: building

Claims: 12

References: Pending verification

Proof: Verification pending

Freshness state: computing

Source paper: Agentic Discovery of Neural Architectures: AIRA-Compose and AIRA-Design

PDF: https://arxiv.org/pdf/2605.15871v1

Repository: https://github.com/facebookresearch/repo

Source count: 4

Coverage: 50%

Last proof check: 2026-05-18T20:27:06.464Z

Signal Canvas receipt window

Ready for execution: Agentic Discovery of Neural Architectures: AIRA-Compose and AIRA-Design

/buildability/agentic-discovery-of-neural-architectures-aira-compose-and-aira-design

Build Nowready

Subject: Agentic Discovery of Neural Architectures: AIRA-Compose and AIRA-Design

Verdict

Build Now

Verdict is Build Now because viability and implementation proof cleared the Wave 1 scaffold thresholds.

Preparing verified analysis

GitHub Code Pulse

No public code linked for this paper yet.

Claim map

Strong 12Mixed 0Weak 0

Evidencepartial
AIRA-Compose uses 11 agents to explore fundamental computational primitives under a 24-hour budget.
Implicationpartial
Directly stated in the abstract with specific numbers.
Verificationpartial
partial
Evidencepartial
AIRA-Compose uses 11 agents to explore fundamental computational primitives under a 24-hour budget.
Implicationpartial
Directly stated in the abstract with specific numbers.
Verificationpartial
partial
Evidencepartial
On downstream tasks, AIRAformer-D and AIRAhybrid-D improve accuracy by 2.4% and 3.8% over Llama 3.2.
Implicationpartial
Explicitly stated with precise percentages in the abstract.
Verificationpartial
partial
Evidencepartial
AIRAformer-C scales 54% and 71% faster than Llama 3.2 and Composer's best Transformer
Implicationpartial
Directly stated with specific percentages in the abstract.
Verificationpartial
partial
Evidencepartial
AIRA-Compose uses 11 agents to explore fundamental computational primitives under a 24-hour budget.
Implicationpartial
Directly stated in the abstract with specific numbers.
Verificationpartial
partial
Evidencepartial
On downstream tasks, AIRAformer-D and AIRAhybrid-D improve accuracy by 2.4% and 3.8% over Llama 3.2.
Implicationpartial
Explicitly stated in the abstract with precise percentages.
Verificationpartial
partial
Evidencepartial
This yields 14 architectures across two families: AIRAformers (Transformer-based) and AIRAhybrids (Transformer-Mamba).
Implicationpartial
Explicitly stated in the abstract.
Verificationpartial
partial
Evidencepartial
AIRA-Design tasks 20 agents with writing novel attention mechanisms for long-range dependencies and high-performing training scripts.
Implicationpartial
Directly stated in the abstract.
Verificationpartial
partial
Evidencepartial
On the Long Range Arena benchmark, agent-designed architectures reach within 2.3% and 2.6% of human state-of-the-art on document matching and text classification.
Implicationpartial
Explicitly stated with specific percentages in the abstract.
Verificationpartial
partial
Evidencepartial
On downstream tasks, AIRAformer-D and AIRAhybrid-D improve accuracy by 2.4% and 3.8% over Llama 3.2.
Implicationpartial
Explicitly stated in the abstract with precise percentages.
Verificationpartial
partial
Evidencepartial
On the Autoresearch benchmark, Greedy Opus 4.5 achieves 0.968 validation bits-per-byte under a fixed time budget, surpassing the published minimum.
Implicationpartial
Directly stated with specific metric in the abstract.
Verificationpartial
partial
Evidencepartial
AIRAformer-C scales 54% and 71% faster than Llama 3.2 and Composer's best Transformer
Implicationpartial
Directly stated in the abstract with specific percentages.
Verificationpartial
partial

Author intelligence and commercialization panels stay hidden until the proof receipt is verified, cites at least 3 references, includes at least 2 sources, and clears 50% coverage. The paper narrative and citation surfaces remain public while verification is pending.

Agentic Discovery of Neural Architectures: AIRA-Compose and AIRA-Design

Use Signal Canvas as the narrative proof surface

Use this Signal Canvas via API or MCP

Signal Canvas proof surface