Unverified50% confidenceBenchmarkExact time
On cross-window, cross-session associative recall tasks, attention mechanisms and LoRA achieve a recall rate of 0.000 by construction, while the fast weight matrix scores 0.62/0.43/0.62 across three seed runs, averaging 0.56; removing the matrix collapses recall to 0.008.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/19/2026
First Seen
Valid until: 1/17/2027
Sources
Related Entities
Related Claims
UnverifiedOn dependency tasks, memory raised the success rate from a leakage-verified baseline of 0.00 to 0.55–1.00.69% similarUnverifiedIn the experiment, the task_score under the shaped reward configuration was only 0.125, compared to 0.417 for the naive reward, with an effect size of d=4.47.68% similarUnverifiedIn a synthetic logistic regression classifier experiment, after 80 steps of gradient-ascent-based repair, the unlearning loss rose from 0.0545 to 0.1649, yet the trigger attack success rate remained 100%.67% similarUnverifiedAfter offloading half of the model layers to system memory, the generation speed often drops sharply from over ten tokens per second to single digits.67% similarUnverified任务的输出熵越低（答案越可预测），推测性解码的加速效益越显著；GSM8K接受率从位置0的0.92衰减至0.53，MTBench从0.78急跌至0.1365% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1067227API
curl https://kongchang.com/api/v1/knowledge/claims/1067227MCP
get_claim(id=1067227)