Beijing Normal University, Zhuhai

Non-prefix KV-cache reuse encodes retrieved chunks separately and concatenates their caches, losing 19% to 100% of context-dependent questions that full prefill answers, with same-kind distractor chunks adding to the loss. Comparing whether a small repair budget should go to distractors or to the answer-bearing chunk, the study finds that restoring all distractors costs nearly a full prefill, while recomputing the answer-bearing chunk beats token-level selectors at a 5% budget in all nine open-domain QA and API-documentation settings.

Non-prefix KV-cache reuse encodes retrieved chunks separately and concatenates their caches, losing 19% to 100% of context-dependent questions that full prefill answers, with same-kind distractor chunks adding to the loss. Comparing whether a small repair budget should go to distractors or to the answer-bearing chunk, the study finds that restoring all distractors costs nearly a full prefill, while recomputing the answer-bearing chunk beats token-level selectors at a 5% budget in all nine open-domain QA and API-documentation settings.

Web platforms must classify new pages before anyone labels them, large language model labels per page are noisy, and links mean opposite things across sites: online stores often join same-category products, while university sites mostly join pages of different classes. ZeroPolar calls the sign with which links enter a prediction structural polarity, reads that sign without true labels from the excess agreement of LLM labels and text-prototype predictions along edges, and applies one signed propagation step. With one GPT-4o label per page, ZeroPolar-T reaches the highest eleven-graph mean, 73.99%, 6.01 points above LLM-BP.

Web platforms must classify new pages before anyone labels them, large language model labels per page are noisy, and links mean opposite things across sites: online stores often join same-category products, while university sites mostly join pages of different classes. ZeroPolar calls the sign with which links enter a prediction structural polarity, reads that sign without true labels from the excess agreement of LLM labels and text-prototype predictions along edges, and applies one signed propagation step. With one GPT-4o label per page, ZeroPolar-T reaches the highest eleven-graph mean, 73.99%, 6.01 points above LLM-BP.

Approximation error in pretrained attention concentrates in a few high-contribution query–key dimensions, and sharing quadratic-kernel coefficients by contribution group lowers frozen-attention relative MSE to 0.432 on Qwen2.5-0.5B and WikiText-103 perplexity from 31.71 to 27.50 after replacing middle-layer attention.

Approximation error in pretrained attention concentrates in a few high-contribution query–key dimensions, and sharing quadratic-kernel coefficients by contribution group lowers frozen-attention relative MSE to 0.432 on Qwen2.5-0.5B and WikiText-103 perplexity from 31.71 to 27.50 after replacing middle-layer attention.