Unverified50% confidenceBenchmarkExact time
At a 4K context, PTTT with two refinement steps evaluates the inner-loop trajectory at matched error 32 times faster than TTT-MLP using chunked approximation.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/30/2026
First Seen
Valid until: 1/28/2027
Sources
Related Entities
Related Claims
UnverifiedOn the Packing 26 problem, the TTT-Discover recipe reached a certified radius sum of 2.635983 within 50 steps, exactly the same as the 2.635983 reported by TTT-Discover, with a target of 2.636, where higher is better.67% similarUnverifiedAcross a range of question sizes spanning 256x, the error judgment rate of a single run is almost unchanged: 0.531 at T=16 and 0.534 at T=4096.64% similarUnverifiedOn the Erdős minimum overlap problem, the TTT-Discover recipe of "eight groups × 64 rollouts plus one LoRA update" reached a certified C₅ upper bound of 0.38094 within 25 steps.64% similarUnverifiedAfter 200 steps, the maximum discrepancy between the corresponding weights of the two runs reaches 0.237.63% similarUnverifiedTTM 在 ETT-1 数据集上的 MSE 仅为 0.363，超越了 TimesFM、PatchTST 等大模型62% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1114897API
curl https://kongchang.com/api/v1/knowledge/claims/1114897MCP
get_claim(id=1114897)