Searcharxiv⌕ Search

arXiv subjects

Shirong Wang

Publications and source records attributed to Shirong Wang.

4 recordsLinked to original sources

Can LLM Agents Organize Their Own Memory?

In most LLM agents, weights are fixed after deployment, so new information must be retained in external memory. Existing approaches often rely on predefined records and fixed organizational rules, which cannot readily accommodate evolving tasks and interactions. This raises the question: can LLM agents organize their own memory using ordinary documents? We introduce self-organized memory (SoMem), a framework that stores memory in linked documents. The agent decides what to summarize, how to group related information, and when to revise existing topics, and every summary it produces stays linked to the original interactions as evidence. The runtime validates updates and automatically derives temporal, relational, and recent-memory views from the organized content. At query time, the agent chooses what to search and read, following document links and inspecting original evidence as needed. On LoCoMo, BEAM-100K, and LongMemEval-S, SoMem outperforms the strongest baseline by 2.66, 3.10, and 2.00 percentage points, respectively. The constructed memory can be reused across models, and additional reasoning effort can further improve answer quality. These findings demonstrate the feasibility of agents organizing and using their own persistent memory through ordinary documents. Code is available at: https://github.com/Qi202/SoMem.

cs.AI↗

GUI-Lens: Enabling Precise GUI Grounding for Off-the-Shelf VLMs via Coarse-to-Fine Cropping

GUI grounding maps natural-language instructions to click locations and is essential for reliable GUI agents. The task remains difficult on high-resolution, densely populated interfaces because a vision-language model (VLM) may recognize a requested control without locating it precisely enough for interaction. Most existing methods provide various forms of localization assistance, but still rely on a direct click prediction, allowing visual ambiguity or an inaccurate initial estimate to propagate to the final result. In this paper, we introduce GUI-Lens, a coarse-to-fine grounding framework that allows a general-purpose VLM to determine the target through active visual observations. Specifically, GUI-Lens extracts OCR text and detected UI components from the screenshot and presents their positions as coordinate references. Using the instruction, the current view, and these references, the VLM selects the region and scale of the next view, which is cropped and enlarged to provide finer visual details. This process continues over successively focused views until the target is determined. Proposed crops and clicks are checked against the instruction throughout the process, and the final local position is mapped back to the original screen coordinates. Experiments on five GUI grounding benchmarks and six general-purpose base models show that GUI-Lens consistently improves grounding accuracy, achieving up to 9.3 times the accuracy of direct grounding on ScreenSpot-Pro.

cs.CV↗

The Python Simulations of Chemistry Framework: 10 years of an open-source quantum chemistry project

Over the past decade, the Python-based Simulations of Chemistry Framework (PySCF) has developed into a widely used open-source platform for electronic structure theory and quantum chemical method development. This article reviews the major advances since the previous overview in 2020, covering new modules and methodology, infrastructure changes, and performance benchmarks.

physics.chem-ph↗

Improving XYG3-type Doubly Hybrid Approximation using Self-Interaction Corrected SCAN Density and Orbitals via the PZ-SIC Framework: the xDH@SCAN(SIC) Approach

XYG3-type doubly hybrid approximations (xDH) have gained a widespread recognition for their accuracy in describing a diverse range of chemical and physical interactions. However, a recent study (J. Phys. Chem. 2021, 12, 800-807) has highlighted the limitation of xDH methods in calculating the dissociation of the NaCl molecule. This issue has been related to the density and orbitals used for evaluating the energy in xDH methods, which are obtained from lower-rung hybrid density functional approximations (DFAs) and display substantial density errors in the dissociation limit. In this work, we systematically investigate the influence of density on several challenging datasets and find that the xDH methods are less sensitive to the density errors compared to semi-local and hybrid DFAs. Furthermore, we demonstrate that the self-interaction corrected SCAN density offers superior accuracy compared to the self-consistent SCAN density and Hartree-Fock (HF) density, as evidenced by the charge analysis on the dissociation of heterodimers, such as NaCl and LiF. Building on these insights, we propose a 5-parameter xDH method using the SCAN density and orbitals corrected by the PZ-SIC scheme. This new xDH@SCAN(SIC) method provides a balanced and accurate description across a wide range of challenging systems.

physics.chem-ph↗