Niantao Xie

Niantao Xie

Brass and Sharifi proved the lower bound 0.832 for the convex form of Lebesgue's universal cover problem by combining geometric estimates with a computer search over placements of a disk, an equilateral triangle, and a regular pentagon. This paper gives a certificate-based reproduction of that computation. The certificate consists of a finite adaptive ledger, a terminal-route replay, three local lower-bound certificate families, compact integrity audits for large tables, and a finite proof-obligation layer connecting the replayed data to the lower-bound statement. Under the stated verification model, acceptance of this finite certificate implies the Brass--Sharifi convex lower bound αcvx $\ge$ 0.832. We claim neither a numerical improvement over the Brass--Sharifi bound nor a nonconvex lower bound; proof-assistant formalization and independent external verification remain outside the present scope.

Lei Wei, Xu Dong, Xiao Peng et al.

Large language models deployed as autonomous agents face critical memory limitations, lacking selective forgetting mechanisms that lead to either catastrophic forgetting at context boundaries or information overload within them. While human memory naturally balances retention and forgetting through adaptive decay processes, current AI systems employ binary retention strategies that preserve everything or lose it entirely. We propose FadeMem, a biologically-inspired agent memory architecture that incorporates active forgetting mechanisms mirroring human cognitive efficiency. FadeMem implements differential decay rates across a dual-layer memory hierarchy, where retention is governed by adaptive exponential decay functions modulated by semantic relevance, access frequency, and temporal patterns. Through LLM-guided conflict resolution and intelligent memory fusion, our system consolidates related information while allowing irrelevant details to fade. Experiments on Multi-Session Chat, LoCoMo, and LTI-Bench demonstrate superior multi-hop reasoning and retrieval with 45\% storage reduction, validating the effectiveness of biologically-inspired forgetting in agent memory systems.

Niantao Xie, Sujian Li, Huiling Ren et al.

Recently, the seq2seq abstractive summarization models have achieved good results on the CNN/Daily Mail dataset. Still, how to improve abstractive methods with extractive methods is a good research direction, since extractive methods have their potentials of exploiting various efficient features for extracting important sentences in one text. In this paper, in order to improve the semantic relevance of abstractive summaries, we adopt the WordNet based sentence ranking algorithm to extract the sentences which are most semantically to one text. Then, we design a dual attentional seq2seq framework to generate summaries with consideration of the extracted information. At the same time, we combine pointer-generator and coverage mechanisms to solve the problems of out-of-vocabulary (OOV) words and duplicate words which exist in the abstractive models. Experiments on the CNN/Daily Mail dataset show that our models achieve competitive performance with the state-of-the-art ROUGE scores. Human evaluations also show that the summaries generated by our models have high semantic relevance to the original text.