arXiv subjects

Yanna Wang

Publications and source records attributed to Yanna Wang.

14 recordsLinked to original sources

arXiv2026

ConvMem: Convolutional Memory for Long-Context Reasoning

While Large Language Models (LLMs) have demonstrated impressive capabilities, they often struggle with extremely long contexts due to fixed context limits. To address this, sequential approaches like MemAgent extend the effective context by reading text in segments and iteratively updating a fixed-size memory. However, this sequential paradigm suffers from high latency and requires costly reinforcement learning (RL) training, which can lead to overfitting on specific datasets. To overcome these limitations, we propose ConvMem, a training-free, highly parallelizable framework that reformulates long-context reasoning as a hierarchical convolution. Inspired by CNNs, ConvMem treats an LLM prompted with a specific query as a convolutional kernel. This kernel summarizes text segments hierarchically, shortening the reasoning path from a linear chain into a logarithmic tree. Specifically, ConvMem integrates \textit{Configurable Strides} and \textit{Skip Connections} to ensure robust evidence capture and propagation, while employing \textit{Multi-Kernel Convolution} to decompose complex queries into disentangled semantic channels. This design not only mitigates error accumulation but also enables massive parallelization across both text segments and reasoning threads. Experiments on RULER-HotpotQA and RULER-2WikiMultiHopQA demonstrate that ConvMem outperforms training-free baselines and avoids the risk of overfitting to parametric priors often observed in RL-trained models on out-of-distribution tasks.

cs.AI↗

arXiv2026

HyMem: Hierarchical Context Management for Long-Horizon Agents via Information Isolation

Large language model (LLM) agents often perform poorly on complex, long-horizon tasks because their context becomes increasingly cluttered over time. As interactions accumulate, detailed execution traces and intermediate outputs dominate the context, making it difficult for the model to retain and use high-level planning information. Most existing methods address this issue through compression or retrieval applied to a single, flat context, which does not clearly separate different types of context information and often leads to degraded reasoning. To address this challenge, we propose HyMem, a hierarchical framework that explicitly separates the agent's context into distinct functional layers. HyMem organizes context by function to separate high-level planning from execution and complex analysis. Its isolated reasoning module handles complex subtasks without adding intermediate reasoning traces to the persistent planning context, while its memory management module preserves task progress across context refreshes through structured summaries. These components reduce redundant context accumulation, retain task-critical information, and support coherent long-horizon reasoning within a limited context window. Experiments on GAIA and Browsecomp-plus show that, with DeepSeek-V4, HyMem achieves average Pass@1 scores of 66.7% and 61.3%, outperforming the strongest baseline by 6.1 and 4.7 percentage points, respectively. Further analysis indicates that HyMem effectively controls the growth of the reasoning context, allowing the model to maintain focus and accuracy across complex, long-horizon tasks.

cs.AI↗

arXiv2026

On distance spectral radius of power hypertrees with given number of pendant paths of fixed length

The distance spectral radius of a connected hypergraph is the largest eigenvalue of the distance matrix of the hypergraph. A pendant path of length l with l greater than or equal to 1 in a hypergraph G at vertex v sub l plus 1 is a path consisting of vertices and edges v1 e1 v2 up to vl el v(l+1). The vertex v(l+1) has degree at least 2, vertex v1 has degree 1, and each vertex vi has degree 2 for i from 2 to l. Every vertex belonging to edge ei except vi and v(i+1) has degree 1 for all i from 1 to l. We find the unique hypertree that maximizes or minimizes the distance spectral radius among all r-th power hypertrees with m edges and k pendant paths of length l, where r is no less than 3, k and l are no less than 1, and the product of k and l is smaller than m.

math.CO↗

arXiv2026

Self-Guided Function Calling in Large Language Models via Stepwise Experience Recall

Function calling enables large language models (LLMs) to interact with external systems by leveraging tools and APIs. When faced with multi-step tool usage, LLMs still struggle with tool selection, parameter generation, and tool-chain planning. Existing methods typically rely on manually designing task-specific demonstrations, or retrieving from a curated library. These approaches demand substantial expert effort and prompt engineering becomes increasingly complex and inefficient as tool diversity and task difficulty scale. To address these challenges, we propose a self-guided method, Stepwise Experience Recall (SEER), which performs fine-grained, stepwise retrieval from a continually updated experience pool. Instead of relying on static or manually curated library, SEER incrementally augments the experience pool with past successful trajectories, enabling continuous expansion of the pool and improved model performance over time. Evaluated on the ToolQA benchmark, SEER achieves an average improvement of 6.1% on easy and 4.7% on hard questions. We further test SEER on $τ$-bench, which includes two real-world domains. Powered by Qwen2.5-7B and Qwen2.5-72B models, SEER demonstrates substantial accuracy gains of 7.44% and 23.38%, respectively.

cs.CL↗

arXiv2026

Coarse-to-Fine Grounded Memory for LLM Agent Planning

Recent advancements in Large Language Models (LLMs) have driven growing interest in LLM-based agents for complex planning tasks. To avoid costly agent training, many studies adopted memory mechanism that enhances LLM with offline experiences or online trajectory analysis. However, existing works focus on single-granularity memory derived from dynamic environmental interactions, which are inherently constrained by the quality of the collected experiences. This limitation, in turn, constrain the diversity of knowledge and the flexibility of planning. We propose Coarse-to-Fine Grounded Memory (\Ours{}), a novel framework that grounds coarse-to-fine memories with LLM, thereby fully leverage them for flexible adaptation to diverse scenarios. \Ours{} grounds environmental information into coarse-grained focus points to guide experience collection in training tasks, followed by grounding of actionable hybrid-grained tips from each experience. At inference, \Ours{} retrieves task-relevant experiences and tips to support planning. When facing environmental anomalies, the LLM grounds the current situation into fine-grained key information, enabling flexible self-QA reflection and plan correction.

cs.AI↗

arXiv2025

Empowering LLMs with Parameterized Skills for Adversarial Long-Horizon Planning

Recent advancements in Large Language Models(LLMs) have led to the development of LLM-based AI agents. A key challenge is the creation of agents that can effectively ground themselves in complex, adversarial long-horizon environments. Existing methods mainly focus on (1) using LLMs as policies to interact with the environment through generating low-level feasible actions, and (2) utilizing LLMs to generate high-level tasks or language guides to stimulate action generation. However, the former struggles to generate reliable actions, while the latter relies heavily on expert experience to translate high-level tasks into specific action sequences. To address these challenges, we introduce the Plan with Language, Act with Parameter (PLAP) planning framework that facilitates the grounding of LLM-based agents in long-horizon environments. The PLAP method comprises three key components: (1) a skill library containing environment-specific parameterized skills, (2) a skill planner powered by LLMs, and (3) a skill executor converting the parameterized skills into executable action sequences. We implement PLAP in MicroRTS, a long-horizon real-time strategy game that provides an unfamiliar and challenging environment for LLMs. The experimental results demonstrate the effectiveness of PLAP. In particular, GPT-4o-driven PLAP in a zero-shot setting outperforms 80% of baseline agents, and Qwen2-72B-driven PLAP, with carefully crafted few-shot examples, surpasses the top-tier scripted agent, CoacAI. Additionally, we design comprehensive evaluation metrics and test 6 closed-source and 2 open-source LLMs within the PLAP framework, ultimately releasing an LLM leaderboard ranking long-horizon skill planning ability. Our code is available at https://github.com/AI-Research-TeamX/PLAP.

cs.CL↗

arXiv2025

Strategy-Augmented Planning for Large Language Models via Opponent Exploitation

Efficiently modeling and exploiting opponents is a long-standing challenge in adversarial domains. Large Language Models (LLMs) trained on extensive textual data have recently demonstrated outstanding performance in general tasks, introducing new research directions for opponent modeling. Some studies primarily focus on directly using LLMs to generate decisions based on the elaborate prompt context that incorporates opponent descriptions, while these approaches are limited to scenarios where LLMs possess adequate domain expertise. To address that, we introduce a two-stage Strategy-Augmented Planning (SAP) framework that significantly enhances the opponent exploitation capabilities of LLM-based agents by utilizing a critical component, the Strategy Evaluation Network (SEN). Specifically, in the offline stage, we construct an explicit strategy space and subsequently collect strategy-outcome pair data for training the SEN network. During the online phase, SAP dynamically recognizes the opponent's strategies and greedily exploits them by searching best response strategy on the well-trained SEN, finally translating strategy to a course of actions by carefully designed prompts. Experimental results show that SAP exhibits robust generalization capabilities, allowing it to perform effectively not only against previously encountered opponent strategies but also against novel, unseen strategies. In the MicroRTS environment, SAP achieves a $85.35\%$ performance improvement over baseline methods and matches the competitiveness of reinforcement learning approaches against state-of-the-art (SOTA) rule-based AI. Our code is available at https://github.com/hsushuai/SAP.

cs.AI↗

arXiv2025

TUMS: Enhancing Tool-use Abilities of LLMs with Multi-structure Handlers

Recently, large language models(LLMs) have played an increasingly important role in solving a wide range of NLP tasks, leveraging their capabilities of natural language understanding and generating. Integration with external tools further enhances LLMs' effectiveness, providing more precise, timely, and specialized responses. However, LLMs still encounter difficulties with non-executable actions and improper actions, which are primarily attributed to incorrect parameters. The process of generating parameters by LLMs is confined to the tool level, employing the coarse-grained strategy without considering the different difficulties of various tools. To address this issue, we propose TUMS, a novel framework designed to enhance the tool-use capabilities of LLMs by transforming tool-level processing into parameter-level processing. Specifically, our framework consists of four key components: (1) an intent recognizer that identifies the user's intent to help LLMs better understand the task; (2) a task decomposer that breaks down complex tasks into simpler subtasks, each involving a tool call; (3) a subtask processor equipped with multi-structure handlers to generate accurate parameters; and (4) an executor. Our empirical studies have evidenced the effectiveness and efficiency of the TUMS framework with an average of 19.6\% and 50.6\% improvement separately on easy and hard benchmarks of ToolQA, meanwhile, we demonstrated the key contribution of each part with ablation experiments, offering more insights and stimulating future research on Tool-augmented LLMs.

cs.CL↗

arXiv2023

A proof of a conjecture on the distance spectral radius

A cactus is a connected graph in which any two cycles have at most one common vertex. We determine the unique graph that maximizes the distance spectral radius over all cacti with fixed numbers of vertices and cycles, and thus prove a conjecture on the distance spectral radius of cacti in [S.S. Bose, M. Nath, S. Paul, On the distance spectral radius of cacti, Linear Algebra Appl. 437 (2012) 2128--2141]. We prove the result in the context of hypertrees.

math.CO↗

arXiv2020

DGST : Discriminator Guided Scene Text detector

Scene text detection task has attracted considerable attention in computer vision because of its wide application. In recent years, many researchers have introduced methods of semantic segmentation into the task of scene text detection, and achieved promising results. This paper proposes a detector framework based on the conditional generative adversarial networks to improve the segmentation effect of scene text detection, called DGST (Discriminator Guided Scene Text detector). Instead of binary text score maps generated by some existing semantic segmentation based methods, we generate a multi-scale soft text score map with more information to represent the text position more reasonably, and solve the problem of text pixel adhesion in the process of text extraction. Experiments on standard datasets demonstrate that the proposed DGST brings noticeable gain and outperforms state-of-the-art methods. Specifically, it achieves an F-measure of 87% on ICDAR 2015 dataset.

cs.CV↗

arXiv2020

Multi-Fusion Chinese WordNet (MCW) : Compound of Machine Learning and Manual Correction

Princeton WordNet (PWN) is a lexicon-semantic network based on cognitive linguistics, which promotes the development of natural language processing. Based on PWN, five Chinese wordnets have been developed to solve the problems of syntax and semantics. They include: Northeastern University Chinese WordNet (NEW), Sinica Bilingual Ontological WordNet (BOW), Southeast University Chinese WordNet (SEW), Taiwan University Chinese WordNet (CWN), Chinese Open WordNet (COW). By using them, we found that these word networks have low accuracy and coverage, and cannot completely portray the semantic network of PWN. So we decided to make a new Chinese wordnet called Multi-Fusion Chinese Wordnet (MCW) to make up those shortcomings. The key idea is to extend the SEW with the help of Oxford bilingual dictionary and Xinhua bilingual dictionary, and then correct it. More specifically, we used machine learning and manual adjustment in our corrections. Two standards were formulated to help our work. We conducted experiments on three tasks including relatedness calculation, word similarity and word sense disambiguation for the comparison of lemma's accuracy, at the same time, coverage also was compared. The results indicate that MCW can benefit from coverage and accuracy via our method. However, it still has room for improvement, especially with lemmas. In the future, we will continue to enhance the accuracy of MCW and expand the concepts in it.

cs.CL↗

arXiv2019

Selective Feature Connection Mechanism: Concatenating Multi-layer CNN Features with a Feature Selector

Different layers of deep convolutional neural networks(CNNs) can encode different-level information. High-layer features always contain more semantic information, and low-layer features contain more detail information. However, low-layer features suffer from the background clutter and semantic ambiguity. During visual recognition, the feature combination of the low-layer and high-level features plays an important role in context modulation. If directly combining the high-layer and low-layer features, the background clutter and semantic ambiguity may be caused due to the introduction of detailed information. In this paper, we propose a general network architecture to concatenate CNN features of different layers in a simple and effective way, called Selective Feature Connection Mechanism (SFCM). Low-level features are selectively linked to high-level features with a feature selector which is generated by high-level features. The proposed connection mechanism can effectively overcome the above-mentioned drawbacks. We demonstrate the effectiveness, superiority, and universal applicability of this method on multiple challenging computer vision tasks, including image classification, scene text detection, and image-to-image translation.

cs.CV↗

arXiv2013

A note on skew spectrum of graphs

We give some properties of skew spectrum of a graph, especially, we answer negatively a problem concerning the skew characteristic polynomial and matching polynomial in [M. Cavers et al., Skew-adjacency matrices of graphs, Linear Algebra Appl. 436 (2012) 4512--4529].

math.CO↗

arXiv2012

A note on the minimum skew rank of a graph

The minimum skew rank $mr^{-}(\mathbb{F},G)$ of a graph $G$ over a field $\mathbb{F}$ is the smallest possible rank among all skew symmetric matrices over $\mathbb{F}$, whose ($i$,$j$)-entry (for $i\neq j$) is nonzero whenever $ij$ is an edge in $G$ and is zero otherwise. We give some new properties of the minimum skew rank of a graph, including a characterization of the graphs $G$ with cut vertices over the infinite field $\mathbb{F}$ such that $mr^{-}(\mathbb{F},G)=4$, determination of the minimum skew rank of $k$-paths over a field $\mathbb{F}$, and an extending of an existing result to show that $mr^{-}(\mathbb{F},G)=2match(G)=MR^{-}(\mathbb{F},G)$ for a connected graph $G$ with no even cycles and a field $\mathbb{F}$, where $match(G)$ is the matching number of $G$, and $MR^{-}(\mathbb{F},G)$ is the largest possible rank among all skew symmetric matrices over $\mathbb{F}$.

math.CO↗