{"schedule": {"version": "0.7", "base_url": "https://pretalx.sciwork.dev/sw26/schedule/", "conference": {"acronym": "sw26", "title": "sciwork 2026", "start": "2026-12-05", "end": "2026-12-05", "daysCount": 1, "timeslot_duration": "00:05", "rooms": [{"name": "403", "guid": null, "description": null, "capacity": 50}, {"name": "402", "guid": null, "description": null, "capacity": 50}], "days": [{"index": 1, "date": "2026-12-05", "day_start": "2026-12-05T04:00:00+00:00", "day_end": "2026-12-06T03:59:00+00:00", "rooms": {"403": [{"id": 113, "guid": "ec34087e-5630-5e19-b53d-1436eabd2d07", "logo": "", "date": "2026-12-05T09:40:00+00:00", "start": "09:40", "duration": "00:30", "room": "403", "slug": "sw26-113-on-the-tenth-anniversary-of-alphago-an-introduction-to-reinforcement-learning-and-planning", "url": "https://pretalx.sciwork.dev/sw26/talk/QEBSV8/", "title": "On the Tenth Anniversary of AlphaGo: An Introduction to Reinforcement Learning and Planning", "subtitle": "", "track": null, "type": "Talk", "language": "en", "abstract": "Ten years after AlphaGo demonstrated the remarkable potential of combining learning and search, reinforcement learning has become an important framework for building agents that learn to make sequential decisions.\r\n\r\nIn this talk, we will give an introduction to the core ideas behind reinforcement learning and planning. We will start from rewards and the basic formulation of sequential decision-making, then introduce value functions, Q-value functions, and policies. From there, we will discuss policy evaluation, policy iteration, and policy-based methods, building an intuitive picture of how an agent can learn to improve its behavior through interaction and experience.\r\n\r\nFinally, we will move from learning to planning and introduce Monte Carlo Tree Search (MCTS), one of the key components behind AlphaGo. By connecting these concepts, the talk aims to provide a conceptual roadmap for understanding how reinforcement learning and search algorithm can work together to solve complex decision-making problems.", "description": "Outline:\r\n- AlphaGo in historical context\r\n- Reinforcement learning concepts: Rewards, value functions, Q-values policy evaluation, policy iteration, and policy method\r\n- Introduction to planning and Monte Carlo Tree Search (MCTS).\r\n\r\nAudience:\r\n- Software engineers\r\n- Data scientists\r\n- Researchers\r\n- Students\r\n\r\nPrerequisites:\r\n- Basic machine learning knowledge is helpful, but no prior experience with reinforcement learning or MCTS is required.", "recording_license": "", "do_not_record": false, "persons": [{"id": 73, "code": "ZXHZQC", "public_name": "Sheng-Wen Cheng", "biography": "Sheng-Wen (Colin) Cheng is a roboticist and software engineer who builds low-level software at the intersection of robotics, control engineering, and real-time systems.\r\n\r\nHe is also an active open-source contributor and has presented his work at leading open-source events, including Open Source Summit organized by the Linux Foundation.\r\n\r\nFor more information, please visit shengwen-tw.github.io", "answers": []}], "links": [], "attachments": [], "answers": []}, {"id": 116, "guid": "24d786dd-ed24-5f8e-8beb-829ed45a8ef2", "logo": "", "date": "2026-12-05T10:20:00+00:00", "start": "10:20", "duration": "00:30", "room": "403", "slug": "sw26-116-pyvista-on-webassembly-enabling-serverless-3d-visualization", "url": "https://pretalx.sciwork.dev/sw26/talk/8HGEEN/", "title": "PyVista on WebAssembly: Enabling Serverless 3D Visualization", "subtitle": "", "track": null, "type": "Talk", "language": "en", "abstract": "PyVista is a VTK-based Python library for 3D data visualization. In this talk, we introduce the technology stack, implementation details, and performance characteristics of `pyvista-wasm`, which enables PyVista to run entirely on WebAssembly (WASM). By combining Pyodide and VTK-WASM, we can deliver full-featured 3D scientific visualization directly in the browser\u2014without requiring a backend server. We will share hands-on technical insights from development alongside live browser-based demos.\r\n\r\n---\r\n\r\n**Background and Motivation**\r\n\r\nHistorically, scientific visualization has relied on local desktop environments or heavy server-side rendering pipelines. With the rise of WebAssembly, heavyweight scientific libraries like VTK are following NumPy and SciPy into browser environments, opening up entirely new architectural possibilities.\r\n\r\n**Talk Outline (30 mins)**\r\n\r\n1. **Overview of PyVista and WASM (5 mins)**\r\n* What is PyVista?\r\n* Why WASM? What changes for scientific workflows?\r\n\r\n\r\n2. **Technology Stack & Implementation Details (12 mins)**\r\n* Emscripten builds of VTK\r\n* Running PyVista on Pyodide\r\n* Integrating with WebGL rendering\r\n* Challenges and solutions in local development and CI pipelines\r\n\r\n\r\n3. **In-Browser Live Demo (8 mins)**\r\n* Running inside JupyterLite\r\n* Interactive 3D mesh manipulation\r\n\r\n\r\n4. **Performance & Limitations (5 mins)**\r\n* Performance comparison: WASM vs. native\r\n* Current WASM constraints and practical workarounds\r\n* Future roadmap", "description": "", "recording_license": "", "do_not_record": false, "persons": [{"id": 51, "code": "KK7KX9", "public_name": "Tetsuo Koyama", "biography": "I am an active contributor to PyVista\u2014a VTK-based Python library for 3D visualization\u2014regularly submitting pull requests for bug fixes and feature enhancements. Expanding on this work, I joined the `pyvista-wasm` project to bring PyVista to WebAssembly, where I now serve as one of the lead developers driving its architecture and implementation. Although I have been involved for less than a year, I have spearheaded core technical efforts, including establishing the build pipeline using Pyodide and VTK-WASM and integrating in-browser rendering. All of this work is done independently as an open-source contributor in my personal time.", "answers": []}], "links": [], "attachments": [], "answers": []}, {"id": 114, "guid": "123e9065-adb8-5f1c-9e91-d8eba0b6b76c", "logo": "", "date": "2026-12-05T11:00:00+00:00", "start": "11:00", "duration": "00:30", "room": "403", "slug": "sw26-114-beyond-calling-blas-implementing-and-optimizing-matrix-multiplication", "url": "https://pretalx.sciwork.dev/sw26/talk/S97GU7/", "title": "Beyond Calling BLAS: Implementing and Optimizing Matrix Multiplication", "subtitle": "", "track": null, "type": "Talk", "language": "en", "abstract": "Matrix multiplication may look like three nested loops or a single BLAS call, but implementing it efficiently in an array library requires more than a fast computational kernel. Shapes, strides, broadcasting, and operand reuse can change which execution strategy works best\u2014even when the arithmetic workload remains the same.\r\n\r\nThis talk explores the implementation and optimization of matrix multiplication in Solvcon\u2019s SimpleArray. We follow the path from direct computation to layout-aware execution, covering BLAS-compatible layouts, signed-stride operands, and batched matrix multiplication. Particular attention is given to small-matrix execution, reuse of broadcast operands, and the trade-offs between complete packing and streamed packing with reusable scratch storage. We also discuss how separating operation semantics, execution selection, and computation makes these optimizations easier to integrate and maintain.\r\n\r\nTo evaluate the implementation, we briefly demonstrate a Qt Benchmark Inspector and explain how identical operands, numerical comparisons, and balanced timing help assess performance improvements. The central lesson is that efficient matrix multiplication depends not only on kernel speed, but also on how data layout, reuse, and execution strategy fit together.", "description": "", "recording_license": "", "do_not_record": false, "persons": [{"id": 50, "code": "9WDRZP", "public_name": "Chun-Shih Chang", "biography": "Chun-Shih Chang works in Acadamic Sininca. He is interested in machine learning, high performance computing and C++ software development.", "answers": []}], "links": [], "attachments": [], "answers": []}, {"id": 131, "guid": "d8af7bd4-ff0e-5b96-a80b-3a2a9098bf7a", "logo": "", "date": "2026-12-05T11:30:00+00:00", "start": "11:30", "duration": "01:20", "room": "403", "slug": "sw26-131-lunch", "url": "https://pretalx.sciwork.dev/sw26/talk/XRJ7SE/", "title": "Lunch", "subtitle": "", "track": null, "type": "Non-Talk", "language": "en", "abstract": "Lunch", "description": "", "recording_license": "", "do_not_record": false, "persons": [], "links": [], "attachments": [], "answers": []}, {"id": 132, "guid": "9434e5f7-84a2-5844-81c1-815808cccd39", "logo": "", "date": "2026-12-05T12:50:00+00:00", "start": "12:50", "duration": "00:50", "room": "403", "slug": "sw26-132-project-sharing", "url": "https://pretalx.sciwork.dev/sw26/talk/NBDZ3J/", "title": "Project Sharing", "subtitle": "", "track": null, "type": "Non-Talk", "language": "en", "abstract": "Project Sharing", "description": "", "recording_license": "", "do_not_record": false, "persons": [], "links": [], "attachments": [], "answers": []}, {"id": 123, "guid": "c7e68377-0486-5df5-98f6-d8da98751903", "logo": "", "date": "2026-12-05T13:50:00+00:00", "start": "13:50", "duration": "00:30", "room": "403", "slug": "sw26-123-werilog-toward-a-bidirectional-coding-architecture", "url": "https://pretalx.sciwork.dev/sw26/talk/STY3KU/", "title": "Werilog: Toward a Bidirectional Coding Architecture", "subtitle": "", "track": null, "type": "Talk", "language": "en", "abstract": "Werilog is a browser-based environment for Hardware Description Languages that removes the \"local toolchain tax\" of hardware design \u2014 no installs, no simulator setup, just coding in the browser.\r\n\r\nAt first glance, Werilog looks like another EDA tool: write Verilog, see it simulate. But the real question we're exploring is the opposite direction \u2014 can you draw a diagram and get working codes back out? Most tools only go one way: code in, diagram out, read-only. We're building toward a loop where the diagram is not just a visualization but an editable, valid representation of the code itself \u2014 and where structured, regular designs (like datapaths and control logic) can be constructed visually and compiled down to correct HDL.\r\n\r\nThis talk covers why we think bidirectionality matters for how people learn and design hardware, what we've built so far (live HDL-to-diagram visualization, in-browser simulation, an early diagram-to-HDL path for structural designs), the architectural choices and constraints that shaped it, and where the project is headed \u2014 including where AI assistance fits into closing the loop.\r\n\r\nThis talk covers why we think bidirectionality matters for how people learn and design hardware, what we've built so far (live HDL-to-diagram visualization, in-browser simulation, an early diagram-to-HDL path for structural designs), and the architectural choices and constraints that shaped it.", "description": "", "recording_license": "", "do_not_record": false, "persons": [{"id": 88, "code": "C8BJ9B", "public_name": "Sivan", "biography": "", "answers": []}], "links": [], "attachments": [], "answers": []}, {"id": 126, "guid": "16ea60cb-9fd1-5813-b847-0d10096f64cb", "logo": "", "date": "2026-12-05T14:30:00+00:00", "start": "14:30", "duration": "00:30", "room": "403", "slug": "sw26-126-shall-we-test-bbgo-", "url": "https://pretalx.sciwork.dev/sw26/talk/MMHXE3/", "title": "Shall We Test? BBGO \u53ef\u6e2c\u8a66\u4ea4\u6613\u7b56\u7565\u958b\u767c", "subtitle": "", "track": null, "type": "Talk", "language": "en", "abstract": "\u6211\u5011\u6703\u5728\u9019\u5834 talk \u88e1\u6703\u4ee5 bbgo \u70ba\u57fa\u790e\uff0c\u5e0c\u671b\u4ee5\u6df1\u5165\u6dfa\u51fa\u7684\u65b9\u5f0f\uff0c\u4ecb\u7d39\u4ee5\u4e0b\u4e3b\u984c\uff1a\r\n- \u56de\u6e2c\u5f15\u64ce\u7684\u5143\u4ef6\u8207\u958b\u767c\r\n- \u53ef\u6e2c\u8a66\u4ea4\u6613\u7b56\u7565\u958b\u767c\r\n\u4f46\u4e0d\u6703\u5305\u542b\uff1a\r\n- \u4ea4\u6613\u7b56\u7565\u76f8\u95dc\u6307\u6a19\u4ecb\u7d39\u8207\u512a\u5316", "description": "\u6574\u5929\u770b\u8457\u4ea4\u6613\u6a19\u7684\u7684 k \u7dda\u4e0a\u4e0a\u4e0b\u4e0b\u8df3\u52d5\uff0c\u8166\u6d77\u4e2d\u662f\u5426\u5ffd\u7136\u5192\u51fa\u4e86\u4e00\u500b\u8072\u97f3\uff1a\u6211\u597d\u50cf\u770b\u51fa\u9019\u5e02\u5834\u7684\u8108\u52d5\u4e86\uff01\r\n\r\n\u6709\u8457\u7121\u6578\u4ea4\u6613\u7d93\u9a57\u7684\u4f60\uff0c\u5df2\u7d93\u5920\u51b7\u975c\u5230\u4e0d\u6703\u99ac\u4e0a\u6253\u958b\u96fb\u8166\u4e0b\u55ae all in\u3002\r\n\u8eab\u70ba\u4e00\u500b\u597d\u7684\u4ea4\u6613\u8005\uff0c\u4f60\u77e5\u9053\u61c9\u8a72\u5148\u5c0d\u4f60\u7684\u4ea4\u6613\u7b56\u7565\u9032\u884c\u6e2c\u8a66\uff0c\r\n\u90a3\u4e00\u500b\u53ef\u4ee5\u4e1f\u9032 bbgo \u56de\u6e2c\u5f15\u64ce\u6e2c\u8a66\u7684\u4ea4\u6613\u7b56\u7565\u53c8\u8a72\u662f\u4ec0\u9ebc\u6a23\u5b50\u5462\uff1f\r\n\r\nbbgo \u662f\u4e00\u500b\u4ee5 Go \u958b\u767c\u7684\u3001\u4ee5\u52a0\u5bc6\u8ca8\u5e63\u5e02\u5834\u70ba\u4e3b\u7684\u4ea4\u6613\u7b56\u7565\u958b\u767c\u6846\u67b6\uff1a\r\nhttps://github.com/c9s/bbgo\r\n\u9019\u5834 talk \u6703\u4ee5 bbgo \u70ba\u57fa\u790e\uff0c\u4ecb\u7d39\u4ea4\u6613\u7b56\u7565\u56de\u6e2c\u7684\u4e00\u4e9b\u57fa\u790e\u6982\u5ff5\u3002\r\n\u66f4\u91cd\u8981\u7684\u662f\uff0c\u5728\u4e00\u500b\u50cf bbgo \u9019\u6a23\u6709\u8457\u9ad8\u5ea6\u81ea\u7531\u5ea6\u7684\u4ea4\u6613\u6846\u67b6\u88e1\uff0c\r\n\u4e00\u500b\"\u53ef\u6e2c\u8a66\"\u7684\u4ea4\u6613\u7b56\u7565\u5728\u958b\u767c\u4e0a\u61c9\u8a72\u8981\u6ce8\u610f\u7684\u4e8b\u60c5\u6709\u54ea\u4e9b\u3002\r\n\r\n\u5e0c\u671b\u53ef\u4ee5\u5728\u9019\u5834 talk \u88e1\u70ba\u5927\u5bb6\u5e36\u4f86\u4e00\u4e9b\u65b0\u77e5\u8207\u60f3\u6cd5\u3002\r\n\u5c31\u7b97\u4e0d\u6703\u6539\u7528 bbgo \u9032\u884c\u4ea4\u6613\uff0c\u5c0d\u65bc\u76ee\u524d\u6b63\u5728\u4f7f\u7528\u7684\u4ea4\u6613\u6846\u67b6\uff0c\r\n\u88e1\u9762\u7684\u76f8\u95dc\u8a2d\u8a08\u8207\u9650\u5236\u4e5f\u6703\u6709\u9032\u4e00\u6b65\u7684\u7406\u89e3\u3002", "recording_license": "", "do_not_record": false, "persons": [{"id": 27, "code": "QC9ALU", "public_name": "Dboy Liao", "biography": "\u76ee\u524d\u4efb\u8077\u65bc MaiCoin \u96c6\u5718\uff0c\u4e3b\u8981\u5de5\u4f5c\u5167\u5bb9\u96c6\u4e2d\u5728\u4ea4\u6613\u7b56\u7565\u958b\u767c\uff0c\u4e5f\u662f bbgo \u4e3b\u8981\u8ca2\u737b\u8005\u4e4b\u4e00\u3002\r\n\u8208\u8da3\u662f\u8b93\u5beb\u5728\u7d19\u4e0a\u7684\u7b97\u5f0f\u53ef\u4ee5\u8dd1\u5728\u96fb\u8166\u88e1\uff0c\u6240\u4ee5\u9664\u4e86\u4ea4\u6613\u5916\u4e5f\u63a5\u89f8\u904e\u4ee5\u4e0b\u9818\u57df\uff1a\r\n- large scale data engineering\r\n- machine learning\r\n- deep learning\r\n\u6240\u4ee5\u73a9\u904e\u4e00\u4e9b\u50cf\u662f Spark, Kafka, Tensorflow, PyTorch \u4e4b\u985e\u7684\uff0c\r\n\u4e5f\u90fd\u6b61\u8fce\u4f86\u627e\u6211\u9592\u804a\u9019\u4e9b\u6771\u897f\u3002\r\n\u5927\u6982\u5c31\u662f\u500b\u6a23\u6a23\u901a\u6a23\u6a23\u9b06\u7684\u8edf\u9ad4\u5de5\u7a0b\u5e2b ^_<", "answers": []}], "links": [], "attachments": [], "answers": []}, {"id": 133, "guid": "c017ad0d-bd96-52f2-84c6-e14eca6c8fef", "logo": "", "date": "2026-12-05T15:00:00+00:00", "start": "15:00", "duration": "01:00", "room": "403", "slug": "sw26-133-tea-break", "url": "https://pretalx.sciwork.dev/sw26/talk/CSADBJ/", "title": "Tea Break", "subtitle": "", "track": null, "type": "Non-Talk", "language": "en", "abstract": "Tea Break", "description": "", "recording_license": "", "do_not_record": false, "persons": [], "links": [], "attachments": [], "answers": []}, {"id": 115, "guid": "94502a06-67c6-5bff-ab82-ff95ac9420ab", "logo": "", "date": "2026-12-05T16:00:00+00:00", "start": "16:00", "duration": "00:30", "room": "403", "slug": "sw26-115--microgpt---llm---lora--python--llm", "url": "https://pretalx.sciwork.dev/sw26/talk/EFCHZR/", "title": "\u5f9e MicroGPT \u5230\u771f\u5be6\u6a21\u578b:\u62c6\u89e3 LLM \u539f\u7406,\u518d\u7528 LoRA \u5fae\u8abf\u6253\u9020 Python \u6587\u4ef6\u7e41\u4e2d\u7ffb\u8b6f LLM", "subtitle": "", "track": null, "type": "Talk", "language": "en", "abstract": "\u5927\u578b\u8a9e\u8a00\u6a21\u578b\u5c0d\u5927\u591a\u6578\u958b\u767c\u8005\u4f86\u8aaa,\u5e38\u5e38\u662f\u4e00\u500b\u300c\u8f38\u5165\u6587\u5b57\u3001\u5410\u51fa\u7b54\u6848\u300d\u7684\u9ed1\u76d2\u5b50\u3002\u6211\u5011\u6bcf\u5929\u547c\u53eb API\u3001\u4e32\u63a5 Chatbot,\u537b\u5f88\u5c11\u771f\u6b63\u62c6\u958b\u4f86\u770b:\u9019\u4e9b\u6a21\u578b\u5167\u90e8\u5230\u5e95\u5728\u7b97\u4ec0\u9ebc?\u7576\u6211\u5011\u8aaa\u300c\u5fae\u8abf\u4e00\u500b\u6a21\u578b\u300d,\u5be6\u969b\u4e0a\u53c8\u662f\u5728\u8abf\u6574\u4ec0\u9ebc\u3001\u53ef\u80fd\u6703\u8e29\u5230\u54ea\u4e9b\u5751?\u9019\u5834\u5206\u4eab\u60f3\u5f9e\u9019\u5169\u500b\u89d2\u5ea6,\u5e36\u5927\u5bb6\u5f9e\u300c\u61c2\u539f\u7406\u300d\u8d70\u5230\u300c\u771f\u7684\u505a\u5f97\u51fa\u4f86\u300d\u3002\r\n\r\n\u9019\u6b21\u5206\u4eab\u4e3b\u8981\u6703\u5206\u6210\u5169\u500b\u6bb5\u843d\uff1a\r\n\r\n\u7b2c\u4e00\u6bb5\u6703\u7528\u4e00\u500b\u6975\u7c21\u7248\u7684 GPT \u5be6\u4f5c(MicroGPT),\u642d\u914d\u4e00\u4efd\u523b\u610f\u8a2d\u8a08\u904e\u7684\u5c0f\u578b\u6e2c\u8a66\u8cc7\u6599,\u73fe\u5834\u5c55\u793a\u4e00\u500b\u8a9e\u8a00\u6a21\u578b\u5f9e\u300c\u4ec0\u9ebc\u90fd\u4e0d\u61c2\u300d\u5230\u300c\u5b78\u6703\u7279\u5b9a\u898f\u5f8b\u300d\u7684\u5b8c\u6574\u904e\u7a0b\u3002\u6e2c\u8a66\u8cc7\u6599\u4f86\u81ea python-docs-zh-tw \u7684\u8853\u8a9e\u5217\u8868,\u4e26\u523b\u610f\u628a\u539f\u672c\u7684\u547d\u540d\u6a5f\u5668\u6539\u6210\u7ffb\u8b6f\u6a5f\u5668(\u4f8b\u5982\u628a\u300caccess\u300d\u5c0d\u61c9\u5230\u300c\u5b58\u53d6\u300d),\u8b93\u8cc7\u6599\u898f\u6a21\u5c0f\u5230\u53ef\u4ee5\u5728\u4e00\u822c\u7b46\u96fb\u4e0a\u3001\u5e7e\u5206\u9418\u5167\u5b8c\u6210\u8a13\u7df4,\u540c\u6642\u4fdd\u7559\u771f\u5be6\u7684\u8a9e\u8a00\u5c0d\u61c9\u95dc\u4fc2\u3002\u900f\u904e\u9019\u500b\u6975\u7c21\u6a21\u578b,\u6211\u5011\u53ef\u4ee5\u6e05\u695a\u770b\u5230:\u6587\u5b57\u662f\u600e\u9ebc\u88ab\u5207\u6210 token\u3001token \u600e\u9ebc\u8b8a\u6210\u5411\u91cf\u3001\u6ce8\u610f\u529b\u6a5f\u5236\u600e\u9ebc\u6c7a\u5b9a\u300c\u9019\u500b\u8a5e\u8a72\u53c3\u8003\u54ea\u4e9b\u4e0a\u4e0b\u6587\u300d\u3002\u56e0\u70ba\u6a21\u578b\u5920\u5c0f\u3001\u8cc7\u6599\u5920\u5c11,\u6574\u500b\u904e\u7a0b\u662f\u5b8c\u5168\u900f\u660e\u3001\u53ef\u4ee5\u9010\u884c\u7a0b\u5f0f\u78bc\u89e3\u91cb\u7684\u3002\r\n\r\n\u7b2c\u4e8c\u6bb5\u6703\u5be6\u969b\u7528\u958b\u6e90\u9810\u8a13\u7df4\u6a21\u578b,\u5fae\u8abf\u51fa\u4e00\u500b\u300cPython \u6280\u8853\u6587\u4ef6\u82f1\u7ffb\u4e2d\u300d\u7684\u5c08\u7528\u7ffb\u8b6f\u6a21\u578b\u3002\u8a13\u7df4\u8cc7\u6599\u5b8c\u5168\u4f86\u81ea python/python-docs-zh-tw \u9019\u500b\u958b\u6e90\u7ffb\u8b6f\u5c08\u6848:\u900f\u904e\u89e3\u6790\u5c08\u6848\u88e1\u7684 .po \u7ffb\u8b6f\u6a94\u6848,\u628a\u6bcf\u4e00\u7d44\u5df2\u7d93\u7531\u793e\u7fa4\u4eba\u5de5\u7ffb\u8b6f\u3001\u5be9\u6838\u904e\u7684\u82f1\u6587\u539f\u6587\u8207\u7e41\u9ad4\u4e2d\u6587\u8b6f\u6587,\u8f49\u63db\u6210\u8a13\u7df4\u7528\u7684\u683c\u5f0f\u3002\r\n\r\n\u5728\u6a21\u578b\u9078\u64c7\u4e0a,\u6703\u5206\u4eab\u5be6\u969b\u5617\u8a66\u904e\u7684\u5e7e\u500b Hugging Face \u958b\u6e90\u6a21\u578b\u4e26\u4f7f\u7528 LoRA(\u900f\u904e Unsloth \u52a0\u901f)\u9019\u7a2e\u53c3\u6578\u9ad8\u6548\u5fae\u8abf\u6280\u8853,\u53ea\u8abf\u6574\u6a21\u578b\u88e1\u4e00\u5c0f\u90e8\u5206\u53c3\u6578,\u5c31\u80fd\u8b93\u6a21\u578b\u5b78\u6703\u7279\u5b9a\u9818\u57df\u7684\u7ffb\u8b6f\u80fd\u529b,\u800c\u4e0d\u9700\u8981\u91cd\u65b0\u8a13\u7df4\u6574\u500b\u6a21\u578b\uff1b\u4ee5\u53ca\u5728\u6a21\u578b\u8a13\u7df4\u6642\u5be6\u969b\u8e29\u904e\u7684\u5404\u7a2e\u5751\u3002", "description": "", "recording_license": "", "do_not_record": false, "persons": [{"id": 63, "code": "QUQ7EY", "public_name": "Timmy", "biography": "\u8edf\u9ad4\u5de5\u7a0b\u5e2b\uff0c\u5c08\u9577\u65bc\u96f2\u7aef\u7cfb\u7d71\u67b6\u69cb\u8207\u5168\u7aef\u958b\u767c\u3002\u8fd1\u671f\u81f4\u529b\u65bc\u7814\u7a76 LLM \u5e95\u5c64\u539f\u7406\uff0c\u5e0c\u671b\u5c0d\u65e5\u5e38\u5de5\u4f5c\u4e2d\u983b\u7e41\u4f7f\u7528\u7684 AI \u6a21\u578b\u6709\u66f4\u6df1\u5165\u7684\u7406\u89e3\u3002", "answers": []}], "links": [], "attachments": [], "answers": []}], "402": [{"id": 129, "guid": "ed448dfe-3f13-5764-abca-e3ca0985eb90", "logo": "", "date": "2026-12-05T09:00:00+00:00", "start": "09:00", "duration": "00:30", "room": "402", "slug": "sw26-129-registration", "url": "https://pretalx.sciwork.dev/sw26/talk/EJEUHD/", "title": "Registration", "subtitle": "", "track": null, "type": "Non-Talk", "language": "en", "abstract": "Registration", "description": "", "recording_license": "", "do_not_record": false, "persons": [], "links": [], "attachments": [], "answers": []}, {"id": 130, "guid": "2ce53019-556e-5062-9c6c-cf0a28ca4646", "logo": "", "date": "2026-12-05T09:30:00+00:00", "start": "09:30", "duration": "00:10", "room": "402", "slug": "sw26-130-opening", "url": "https://pretalx.sciwork.dev/sw26/talk/HXXWPE/", "title": "Opening", "subtitle": "", "track": null, "type": "Non-Talk", "language": "en", "abstract": "Opening", "description": "", "recording_license": "", "do_not_record": false, "persons": [], "links": [], "attachments": [], "answers": []}, {"id": 122, "guid": "aed67d36-1983-5457-aeda-a038c98bee9a", "logo": "", "date": "2026-12-05T09:40:00+00:00", "start": "09:40", "duration": "00:30", "room": "402", "slug": "sw26-122--ai---mechanistic-interpretability-", "url": "https://pretalx.sciwork.dev/sw26/talk/VTNVUA/", "title": "\u6253\u958b AI \u7684\u9ed1\u76d2\u5b50\uff1a\u6a5f\u5236\u53ef\u89e3\u91cb\u6027\uff08Mechanistic Interpretability\uff09\u7684\u6700\u65b0\u9032\u5c55", "subtitle": "", "track": null, "type": "Talk", "language": "en", "abstract": "\u9019\u5834\u6f14\u8b1b\u5c07\u5e36\u60a8\u7a7f\u900f LLM \u9ed1\u7bb1\uff0c\u63a2\u7d22\u6a5f\u5236\u53ef\u89e3\u91cb\u6027\uff08MI\uff09\u7684\u6700\u65b0\u9032\u5c55\uff0c\u5b83\u672c\u8cea\u4e0a\u662f\u4e00\u9805\u7d50\u5408\u591a\u500b\u7814\u7a76\u9818\u57df\u7684\u6311\u6230\u3002\u6211\u5011\u5c07\u63f4\u5f15\u5404\u500b\u5047\u8aaa\u4f86\u5256\u6790\u6a21\u578b\u5167\u90e8\u7684\u6982\u5ff5\u904b\u4f5c\uff0c\u8a0e\u8ad6\u5982\u4f55\u8fd1\u4f3c\u624b\u8853\u822c\u7cbe\u6e96\u53bb\u89c0\u5bdf\u3001\u64cd\u63a7\u6a21\u578b\uff0c\u4e26\u7406\u6027\u5c55\u671b\u5ef6\u4f38\u554f\u984c\u3002", "description": "\u9019\u5834\u6f14\u8b1b\u5c07\u5e36\u60a8\u7a7f\u900f LLM \u9ed1\u7bb1\uff0c\u63a2\u8a0e\u300c\u6a5f\u5236\u53ef\u89e3\u91cb\u6027\uff08MI\uff09\u300d\u7684\u6700\u65b0\u9032\u5c55\u3002\u9019\u672c\u8cea\u4e0a\u662f\u4e00\u5834\u7d50\u5408\u96fb\u8166\u79d1\u5b78\u3001\u6578\u5b78\u3001\u795e\u7d93\u79d1\u5b78\u8207\u8a9e\u8a00\u5b78\u7684\u8de8\u9818\u57df\u6311\u6230\u3002\r\n\r\n\u9762\u5c0d\u9ad8\u7dad\u7a7a\u9593\u4e2d\u7684\u5e7e\u4f55\u758a\u52a0\u614b\uff08superposition\uff09\uff0c\u7814\u7a76\u793e\u7fa4\u6b63\u5f9e\u591a\u500b\u5c64\u6b21\u63d0\u51fa\u5047\u8aaa\u2014\u2014\u5f9e\u55ae\u4e00\u6982\u5ff5\u7684\u7dda\u6027\u8868\u5fb5\uff08linear representations\uff09\u3001\u4f4e\u7dad\u5b50\u7a7a\u9593\uff08subspaces\uff09\u7684\u5e7e\u4f55\u6d41\u5f62\uff08geometry of concepts\uff1bmanifolds\uff09\uff0c\u5230\u52d5\u614b\u8a08\u7b97\u7684\u795e\u7d93\u8ff4\u8def\uff08circuits\uff09\u8207\u5438\u5f15\u5b50\u5c71\u8c37\uff08attractor basins\uff09\u3002\u9019\u4e9b\u5047\u8aaa\u4e26\u975e\u4e92\u65a5\uff0c\u800c\u662f\u5c64\u5c64\u4ea4\u7e54\u6210\u7406\u89e3\u5fc3\u667a\u7684\u62d3\u64b2\u5730\u5716\u3002\r\n\r\n\u6211\u5011\u5c07\u5256\u6790\u6a21\u578b\u5167\u90e8\u7684\u6982\u5ff5\u904b\u4f5c\uff0c\u5f9e\u300c\u7406\u8ad6\u901a\u5247\u548c\u55ae\u4f8b\u4e8b\u5be6\uff0c\u5728\u6a21\u578b\u5167\u662f\u5982\u4f55\u88ab\u8868\u5fb5\uff08represented\uff09\uff1f\u300d\u3001\u300c\u80fd\u5426\u70ba\u6a21\u578b\u63a8\u7406\u505a\u51fa\u6709\u8a9e\u610f\u7684\u3001\u6982\u5ff5\u5316\u7684\u89e3\u91cb\uff0c\u800c\u975e\u505c\u7559\u5728\u77e9\u9663\u904b\u7b97\uff1f\u751a\u81f3\u8aaa\u660e\u5b83\u8de8\u9818\u57df\u985e\u6bd4\u80fd\u529b\u5982\u6b64\u5f37\u5927\uff1f\u300d\u9019\u6a23\u7684\u554f\u984c\u51fa\u767c\uff0c\u5f15\u7528\u5404\u500b\u5047\u8aaa\u5617\u8a66\u505a\u51fa\u8aaa\u660e\uff0c\u4e5f\u8a0e\u8ad6\u7576\u4e0b\u6280\u8853\u5982\u4f55\u5be6\u73fe\u7cbe\u6e96\u7684\u6a21\u578b\u89c0\u5bdf\u8207\u64cd\u63a7\uff0c\u4e26\u7406\u6027\u53cd\u601d\u76f8\u95dc\u7684\u5ef6\u4f38\u554f\u984c\uff0c\u4f8b\u5982\uff1aLLM \u7684\u7a0b\u5f0f\u958b\u767c\u80fd\u529b\u70ba\u4f55\u5982\u6b64\u5353\u8d8a\u3002", "recording_license": "", "do_not_record": false, "persons": [{"id": 84, "code": "A389Y9", "public_name": "Junli Xiao", "biography": "", "answers": []}], "links": [], "attachments": [], "answers": []}, {"id": 118, "guid": "5649388d-9e5d-59d5-91bf-f17ee7d5cb9d", "logo": "", "date": "2026-12-05T10:20:00+00:00", "start": "10:20", "duration": "00:30", "room": "402", "slug": "sw26-118-ai-agent-x-open-source-compiler", "url": "https://pretalx.sciwork.dev/sw26/talk/88UBCZ/", "title": "AI Agent x Open Source Compiler", "subtitle": "", "track": null, "type": "Talk", "language": "en", "abstract": "\u4e09\u5e74\u524d\uff0c\u807d\u5230\u6709\u4eba\u60f3\u7528 AI \u5e6b GCC \u52a0\u4e0a\u65b0\u7684\u6307\u4ee4\u652f\u63f4\u6642\uff0c\u6211\u7684\u7b2c\u4e00\u53cd\u61c9\u662f\uff1aX! \u7b11\u6b7b!\r\n\r\n\u4e09\u5e74\u5f8c\uff0cAI Agent \u5b8c\u5168\u878d\u5165\u6211\u7684\u65e5\u5e38 compiler \u958b\u767c\u6d41\u7a0b\uff1a\u4fee bug\u3001\u52a0\u6e2c\u8a66\u3001\u7814\u7a76\u6548\u80fd\u554f\u984c\u3001\u5206\u6790 IR \u8207 assembly\u3002\r\n\r\n\u9019\u5834\u6f14\u8b1b\u6703\u5f9e\u4e00\u4f4d Open Source Compiler Developer \u7684\u89d2\u5ea6\uff0c\u5206\u4eab\u5728 GCC\u3001LLVM \u7b49\u5927\u578b\u958b\u6e90\u5c08\u6848\u4e2d\u4f7f\u7528 AI Agent \u7684\u7d93\u9a57\u3002", "description": "", "recording_license": "", "do_not_record": false, "persons": [{"id": 86, "code": "MJCM8Q", "public_name": "Kito Cheng", "biography": "Open Source Compiler Developer from SiFive", "answers": []}], "links": [], "attachments": [], "answers": []}, {"id": 117, "guid": "a8cfcf57-d2a9-56b8-a216-8de03159e472", "logo": "", "date": "2026-12-05T11:00:00+00:00", "start": "11:00", "duration": "00:30", "room": "402", "slug": "sw26-117---codex-", "url": "https://pretalx.sciwork.dev/sw26/talk/BPRMPT/", "title": "\u5f9e\u81ea\u7136\u8a9e\u8a00\u5230\u7a7a\u9593\u97f3\u8a0a\uff1a\u7528 Codex \u6253\u9020\u53ef\u9a57\u8b49\u7684\u700f\u89bd\u5668\u8072\u5b78\u539f\u578b", "subtitle": "", "track": null, "type": "Talk", "language": "en", "abstract": "\u7576 AI Agent \u80fd\u5920\u5354\u52a9\u7522\u751f\u5927\u91cf\u7a0b\u5f0f\u78bc\uff0c\u958b\u767c\u8005\u5982\u4f55\u78ba\u8a8d\u4e00\u500b\u8de8\u8d8a\u5e7e\u4f55\u3001\u8072\u5b78\u8207\u5373\u6642\u4e92\u52d5\u7684\u61c9\u7528\uff0c\u771f\u7684\u7b26\u5408\u539f\u5148\u7684\u8a2d\u8a08\uff1f\r\n\r\n\u3000\u3000\u672c\u6f14\u8b1b\u4ee5 OpenAI Build Week \u671f\u9593\u958b\u767c\u7684 Echo Canvas \u70ba\u6848\u4f8b\uff0c\u5206\u4eab\u5982\u4f55\u8207 Codex \u5354\u4f5c\uff0c\u5c07\u81ea\u7136\u8a9e\u8a00\u63cf\u8ff0\u8207\u53ef\u7de8\u8f2f\u7684\u7a7a\u9593\u5834\u666f\uff0c\u8f49\u63db\u6210\u80fd\u5728\u700f\u89bd\u5668\u4e2d\u64cd\u4f5c\u8207\u8a66\u807d\u7684\u7a7a\u9593\u97f3\u8a0a\u539f\u578b\u3002\u4f7f\u7528\u8005\u53ef\u4ee5\u8abf\u6574\u7246\u9762\u3001\u958b\u53e3\u3001\u97f3\u6e90\u8207\u8046\u807d\u4f4d\u7f6e\uff0c\u89c0\u5bdf\u8072\u97f3\u8def\u5f91\uff0c\u4e26\u6bd4\u8f03\u906e\u853d\u3001\u65e9\u671f\u53cd\u5c04\u8207\u6b98\u97ff\u5e36\u4f86\u7684\u8b8a\u5316\u3002\r\n\r\n\u3000\u3000\u6f14\u8b1b\u5c07\u8aaa\u660e\u5169\u500b\u4e92\u76f8\u914d\u5408\u7684\u8a2d\u8a08\uff1a\u4ee5\u898f\u683c\u3001\u8cc7\u6599\u5951\u7d04\u8207\u6e2c\u8a66\u7d04\u675f Codex \u7684\u958b\u767c\u5de5\u4f5c\uff1b\u5728\u61c9\u7528\u5167\uff0c\u8b93 GPT-5.6 \u8ca0\u8cac\u7de8\u8b6f\u5019\u9078\u5834\u666f\u53ca\u89e3\u91cb\u7d50\u679c\uff0c\u4e26\u900f\u904e\u7d50\u69cb\u8207\u9818\u57df\u9a57\u8b49\uff0c\u628a\u6a21\u578b\u8f38\u51fa\u4ea4\u7d66\u78ba\u5b9a\u6027\u7684\u5e7e\u4f55\u8207\u8072\u5b78\u8a08\u7b97\uff0c\u518d\u7531 Web Audio \u5b8c\u6210\u700f\u89bd\u5668 HRTF \u6e32\u67d3\u3002\r\n\r\n\u3000\u3000\u900f\u904e\u67b6\u69cb\u62c6\u89e3\u8207 Demo\uff0c\u6211\u5011\u5c07\u8a0e\u8ad6\u5982\u4f55\u754c\u5b9a AI \u7684\u8cac\u4efb\u3001\u6aa2\u67e5\u751f\u6210\u7d50\u679c\uff0c\u4ee5\u53ca\u5340\u5206\u81ea\u52d5\u6e2c\u8a66\u8207\u4eba\u8033\u9a57\u6536\u6240\u80fd\u8b49\u660e\u7684\u4e8b\u60c5\u3002\r\n\r\n\u3000\u3000Echo Canvas \u5b9a\u4f4d\u70ba\u7528\u65bc\u7a7a\u9593\u97f3\u8a0a\u539f\u578b\u8a2d\u8a08\u8207\u9810\u8996\u7684\u4e92\u52d5\u5f0f\u8072\u5b78\u8fd1\u4f3c\uff1b\u6f14\u8b1b\u4e5f\u6703\u4ea4\u4ee3\u8fd1\u4f3c\u65b9\u6cd5\u8207\u4f7f\u7528\u9650\u5236\u3002\u807d\u773e\u5c07\u5e36\u8d70\u4e00\u5957\u628a AI \u5354\u4f5c\u3001\u53ef\u9a57\u8b49\u8a08\u7b97\u8207\u4e92\u52d5\u5c55\u793a\u9023\u63a5\u8d77\u4f86\u7684\u5de5\u7a0b\u5be6\u4f5c\u601d\u8def\u3002", "description": "\u672c\u6f14\u8b1b\u4ee5 Echo Canvas \u7684 Build Week \u7248\u672c\u70ba\u4e3b\uff0c\u5f9e\u4e00\u500b\u5177\u9ad4\u554f\u984c\u51fa\u767c\uff1a\u7576\u97f3\u6e90\u79fb\u5230\u7246\u5f8c\uff0c\u6216\u623f\u9593\u6750\u8cea\u6539\u8b8a\u6642\uff0c\u5982\u4f55\u8b93\u4f7f\u7528\u8005\u5728\u700f\u89bd\u5668\u88e1\u770b\u898b\u8a08\u7b97\u8def\u5f91\u3001\u807d\u898b\u6548\u679c\uff0c\u4e26\u7406\u89e3\u7cfb\u7d71\u505a\u4e86\u54ea\u4e9b\u8fd1\u4f3c\uff1f\r\n\r\n\u5167\u5bb9\u5c07\u6db5\u84cb\u5f9e\u9700\u6c42\u5230\u9a57\u8b49\u7684\u958b\u767c\u6d41\u7a0b\uff0c\u5305\u62ec\u5982\u4f55\u628a\u529f\u80fd\u7bc4\u570d\u3001\u8cc7\u6599\u5951\u7d04\u8207\u9a57\u6536\u689d\u4ef6\u4ea4\u7d66 Codex\uff0c\u5982\u4f55\u4ee5\u5e7e\u4f55\u8207\u8072\u5b78\u6e2c\u8a66\u6aa2\u67e5\u5be6\u4f5c\uff0c\u4ee5\u53ca\u5982\u4f55\u4fdd\u7559\u4eba\u5de5\u8a66\u807d\u4f5c\u70ba\u53e6\u4e00\u5c64\u9a57\u8b49\u3002\u5728\u7cfb\u7d71\u67b6\u69cb\u4e0a\uff0c\u5c07\u4ecb\u7d39\u81ea\u7136\u8a9e\u8a00\u5230 SceneSpec \u7684\u9a57\u8b49\u908a\u754c\u3001Worker \u4e2d\u7684\u78ba\u5b9a\u6027\u8a08\u7b97\uff0c\u4ee5\u53ca Web Audio \u97f3\u8a0a\u6e32\u67d3\u7684\u5206\u5de5\uff0c\u4e26\u4ee5\u76f4\u63a5\u8def\u5f91\u906e\u853d\u3001\u958b\u53e3\u611f\u77e5\u8072\u97f3\u50b3\u64ad\u8207\u4e00\u968e\u65e9\u671f\u53cd\u5c04\u8aaa\u660e\u5373\u6642\u4e92\u52d5\u4e0b\u7684\u5de5\u7a0b\u53d6\u6368\u3002\r\n\r\n\u9810\u8a08\u5167\u5bb9\uff1a\r\n- \u554f\u984c\u80cc\u666f\u3001Build Week \u76ee\u6a19\u8207 Echo Canvas \u4f7f\u7528\u60c5\u5883\r\n- \u8207 Codex \u5354\u4f5c\uff1a\u9700\u6c42\u62c6\u89e3\u3001\u8cc7\u6599\u5951\u7d04\u3001\u6e2c\u8a66\u8207\u4fee\u6b63\r\n- GPT \u5834\u666f\u7de8\u8b6f\u3001\u78ba\u5b9a\u6027\u8072\u5b78\u8a08\u7b97\u8207\u700f\u89bd\u5668\u97f3\u8a0a\u67b6\u69cb\r\n- Demo\uff1a\u79fb\u52d5\u97f3\u6e90\u8207\u8046\u807d\u8005\u3001\u8abf\u6574\u5834\u666f\uff0c\u5c0d\u7167\u8072\u97f3\u8def\u5f91\u53ca Raw\uff0fSimulated \u6548\u679c\r\n- \u8fd1\u4f3c\u65b9\u6cd5\u7684\u9650\u5236\u3001\u9a57\u8b49\u76f2\u9ede\u8207\u958b\u767c\u5fc3\u5f97\r\n- Q&A\r\n\r\n\u76ee\u6a19\u807d\u773e\uff1a\u5c0d AI Agent \u8f14\u52a9\u958b\u767c\u3001Web Audio\u3001\u4e92\u52d5\u5a92\u9ad4\u6216\u79d1\u5b78\u8207\u5de5\u7a0b\u8edf\u9ad4\u6709\u8208\u8da3\u7684\u958b\u767c\u8005\u3002\u5177\u5099\u57fa\u672c\u7a0b\u5f0f\u8a2d\u8a08\u6982\u5ff5\u5373\u53ef\uff0c\u4e0d\u9700\u8072\u5b78\u5c08\u696d\u80cc\u666f\uff0c\u4e5f\u4e0d\u9700\u4e8b\u5148\u5b89\u88dd\u5de5\u5177\u3002\r\n\r\n\u9810\u671f\u6536\u7a6b\uff1a\r\n1. \u80fd\u5340\u5206 AI \u751f\u6210\u5167\u5bb9\u3001\u9818\u57df\u9a57\u8b49\u8207\u78ba\u5b9a\u6027\u8a08\u7b97\u5404\u81ea\u61c9\u8ca0\u8cac\u7684\u5de5\u4f5c\u3002\r\n2. \u7406\u89e3\u5982\u4f55\u7528\u898f\u683c\u8207\u6e2c\u8a66\u652f\u63f4 AI \u5354\u4f5c\uff0c\u4e26\u8fa8\u8b58\u81ea\u52d5\u6e2c\u8a66\u672a\u6db5\u84cb\u7684\u611f\u77e5\u9ad4\u9a57\u3002\r\n3. \u8a8d\u8b58\u700f\u89bd\u5668\u7a7a\u9593\u97f3\u8a0a\u539f\u578b\u7684\u57fa\u672c\u67b6\u69cb\uff0c\u4ee5\u53ca\u4e92\u52d5\u5f0f\u8072\u5b78\u8fd1\u4f3c\u7684\u7528\u9014\u8207\u9650\u5236\u3002\r\n\r\n\u6e2c\u8a66\u9023\u7d50\uff1a\r\nhttps://echo-canva.vercel.app/", "recording_license": "", "do_not_record": false, "persons": [{"id": 85, "code": "YUCL9Y", "public_name": "Kevin", "biography": "\u76ee\u524d\u5c31\u8b80\u5927\u5b78\u4e09\u5e74\u7d1a\uff0c\u517c\u5177\u6df7\u97f3\u5de5\u7a0b\u5e2b\u3001\u97f3\u6a02\u88fd\u4f5c\u4eba\u8207\u8edf\u9ad4\u5de5\u7a0b\u5e2b\u80cc\u666f\uff0c\u4e5f\u662f Echo Canvas \u958b\u767c\u8005\u8207 AI \u4eba\u6a5f\u4e92\u52d5\u7814\u7a76\u8005\u3002\u5f9e\u8072\u97f3\u88fd\u4f5c\u7684\u5be6\u52d9\u51fa\u767c\uff0c\u63a2\u7d22\u97f3\u8a0a\u6280\u8853\u3001\u4e92\u52d5\u4ecb\u9762\u8207 AI \u5354\u4f5c\u5982\u4f55\u7d50\u5408\uff0c\u8b93\u5275\u4f5c\u60f3\u6cd5\u8f49\u5316\u70ba\u53ef\u64cd\u4f5c\u3001\u53ef\u9a57\u8b49\u7684\u5de5\u5177\u3002\r\n\r\n\u5728 OpenAI Build Week \u671f\u9593\uff0c\u8207 Codex \u5354\u4f5c\u958b\u767c Echo Canvas\uff0c\u5c07\u81ea\u7136\u8a9e\u8a00\u5834\u666f\u63cf\u8ff0\u3001\u7a7a\u9593\u7de8\u8f2f\u8207\u78ba\u5b9a\u6027\u8072\u5b78\u8a08\u7b97\u6574\u5408\u70ba\u700f\u89bd\u5668\u7a7a\u9593\u97f3\u8a0a\u539f\u578b\u3002\u95dc\u6ce8\u958b\u767c\u8005\u5982\u4f55\u5411 AI \u8868\u9054\u610f\u5716\u3001\u6aa2\u67e5\u751f\u6210\u7d50\u679c\uff0c\u4e26\u7d50\u5408\u5de5\u7a0b\u6e2c\u8a66\u8207\u5be6\u969b\u8046\u807d\u8a55\u4f30\u4f7f\u7528\u9ad4\u9a57\u3002\u5e0c\u671b\u900f\u904e\u8de8\u8d8a\u97f3\u6a02\u88fd\u4f5c\u3001\u8edf\u9ad4\u5de5\u7a0b\u8207\u4eba\u6a5f\u4e92\u52d5\u7684\u5be6\u4f5c\u7d93\u9a57\uff0c\u8207\u793e\u7fa4\u4ea4\u6d41 AI \u8f14\u52a9\u5275\u4f5c\u53ca\u958b\u767c\u7684\u53ef\u80fd\u6027\u8207\u9650\u5236\u3002", "answers": []}], "links": [], "attachments": [], "answers": []}, {"id": 135, "guid": "7efff0d7-59ea-5f51-8afd-2cca65b9119e", "logo": "", "date": "2026-12-05T11:30:00+00:00", "start": "11:30", "duration": "01:20", "room": "402", "slug": "sw26-135-lunch", "url": "https://pretalx.sciwork.dev/sw26/talk/JJXHKS/", "title": "Lunch", "subtitle": "", "track": null, "type": "Non-Talk", "language": "en", "abstract": "Lunch", "description": "", "recording_license": "", "do_not_record": false, "persons": [], "links": [], "attachments": [], "answers": []}, {"id": 111, "guid": "11c4d4b6-c407-5415-8fb4-e2a85eda18b6", "logo": "", "date": "2026-12-05T13:50:00+00:00", "start": "13:50", "duration": "02:40", "room": "402", "slug": "sw26-111--balloon-popping-challenge", "url": "https://pretalx.sciwork.dev/sw26/talk/CPDGQT/", "title": "\u9ad4\u9a57\u706b\u7bad\u5c0e\u63a7 Balloon Popping Challenge", "subtitle": "", "track": null, "type": "Workshop", "language": "en", "abstract": "\u5c0f\u6642\u5019\u662f\u4e0d\u662f\u4e5f\u66fe\u5e7b\u60f3\u904e\u5beb\u7a0b\u5f0f\u63a7\u5236\u706b\u7bad\u5347\u7a7a\uff1f\u73fe\u5728\uff0c\u4f60\u7684 code \u771f\u7684\u80fd\u4e3b\u5bb0\u706b\u7bad\u7684\u98db\u884c\u8ecc\u8de1\uff01\u96a8\u8457\u81fa\u7063\u5165\u8ecc\u706b\u7bad\u8a08\u756b\u7684\u63a8\u9032\uff0c\u6211\u5011\u6b63\u5fb5\u62db\u300c\u5c0e\u822a\u3001\u5236\u5c0e\u8207\u63a7\u5236\uff08GNC\uff09\u300d\u7684\u8de8\u9818\u57df\u8edf\u9ad4\u4eba\u624d\uff0c\u4e26\u5206\u4eab\u570b\u5bb6\u592a\u7a7a\u4e2d\u5fc3\uff08TASA\uff09\u8207\u967d\u660e\u4ea4\u901a\u5927\u5b78\u524d\u77bb\u706b\u7bad\u7814\u7a76\u4e2d\u5fc3\uff08ARRC\uff09\u5171\u540c\u63a8\u52d5\u7684\u300c[\u706b\u7bad\u5c0e\u63a7\u8edf\u9ad4\u7af6\u8cfd\uff08Balloon Popping Challenge\uff09](https://www.tasa.org.tw/en-US/events/detail/273c7841-ce66-400a-a501-1bf0f7fdc17e)\u300d\u3002\r\n\r\n\u9019\u662f\u500b\u7d50\u5408 [Gymnasium](https://github.com/Farama-Foundation/Gymnasium) \u8a13\u7df4\u74b0\u5883\u8207 [ActiveRocketPy](https://github.com/ARRC-Rocket/BalloonPoppingChallenge) \u6a21\u64ec\u7cfb\u7d71(\u57fa\u65bc\u958b\u6e90\u5c08\u6848 [RocketPy](https://github.com/RocketPy-Team/RocketPy)) \u7684\u706b\u7bad\u5c0e\u63a7\u7af6\u8cfd\u3002\u6d3b\u52d5\u4e2d\u6703\u624b\u628a\u624b\u5e36\u9818\u8207\u6703\u8005\u5617\u8a66\u63a7\u5236\u706b\u7bad\u7684\u63a8\u529b\u5411\u91cf\u63a7\u5236\uff08TVC\uff09\u7cfb\u7d71\uff0c\u4e26\u9032\u4e00\u6b65\u63a2\u8a0e\u5982\u4f55\u61c9\u5c0d\u9663\u98a8\u3001\u611f\u6e2c\u5668\u96dc\u8a0a\u7b49\u8907\u96dc\u7684\u771f\u5be6\u60c5\u5883\u3002\r\n\r\n\u300cBe a User -> Be a Contributor\u300d\u662f\u9019\u5834\u5de5\u4f5c\u574a\u7684\u6838\u5fc3\u7cbe\u795e\uff01\u6211\u5011\u975e\u5e38\u9f13\u52f5\u5927\u5bb6\u53c3\u8207\u958b\u6e90\u8ca2\u737b\uff1a\u7121\u8ad6\u4f60\u60f3\u767c\u63ee Computer Science \u80cc\u666f\u5e6b\u5fd9\u6253\u9020\u98db\u884c\u8ecc\u8de1\u7684\u8cc7\u6599\u8996\u89ba\u5316\u5de5\u5177\uff0c\u9084\u662f\u5c0d\u7269\u7406\u5b78\u6709\u71b1\u5ff1\uff0c\u60f3\u767c PR \u6539\u5584\u706b\u7bad Python \u6a21\u7d44\u7684\u6a21\u64ec\u7279\u6027\uff0c\u9019\u88e1\u90fd\u6709\u4f60\u5927\u5c55\u8eab\u624b\u7684\u821e\u53f0\u3002\r\n\r\n\u672c\u5de5\u4f5c\u574a\u7121\u9700\u4e8b\u5148\u5831\u540d\uff0c\u656c\u8acb\u651c\u5e36\u7b46\u96fb\uff0c\u4e00\u8d77\u6210\u70ba\u592a\u7a7a\u958b\u6e90\u5c08\u6848\u7684\u8ca2\u737b\u8005\uff01", "description": "", "recording_license": "", "do_not_record": false, "persons": [{"id": 82, "code": "WFNWXC", "public_name": "William Mou", "biography": "\u71b1\u8877\u65bc\u592a\u7a7a\u8207\u5929\u6587\u9818\u57df\u7684\u8edf\u9ad4\u5de5\u7a0b\u5e2b\u3002\u56e0 NASA Hackathon \u8207\u5925\u4f34\u7d50\u7de3\uff0c\u66fe\u5728\u6e05\u5927\u53c3\u8207\u885b\u661f\u8a08\u756b\u8207\u5929\u6587\u89c0\u6e2c\uff0c\u8fd1\u671f\u66f4\u6295\u5165\u706b\u7bad\u5c0e\u63a7\u8edf\u9ad4\u7af6\u8cfd\u958b\u767c\u3002\u4e0d\u904e\uff0c\u6211\u7684\u672c\u8077\u5176\u5be6\u662f HPC \u53ca RISC-V \u8edf\u9ad4\u6548\u80fd\u5de5\u7a0b\u5e2b\u3002", "answers": []}], "links": [], "attachments": [], "answers": []}, {"id": 134, "guid": "85b8936c-9deb-5497-b7ee-d7cbca07b7f0", "logo": "", "date": "2026-12-05T16:50:00+00:00", "start": "16:50", "duration": "00:10", "room": "402", "slug": "sw26-134-closing", "url": "https://pretalx.sciwork.dev/sw26/talk/DSWD9R/", "title": "Closing", "subtitle": "", "track": null, "type": "Non-Talk", "language": "en", "abstract": "Closing", "description": "", "recording_license": "", "do_not_record": false, "persons": [], "links": [], "attachments": [], "answers": []}]}}]}}}