--- title: "Jie Tang — papers, page 2 · South China University of Technology" description: "Jie Tang, South China University of Technology: 35 papers on Elamur Research (2021–2026), h-index 82 — affiliations, topics, co-authors…" canonical: "https://research.elamur.ai/free/en/authors/jie-tang/page/2/" --- [Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/authors/jie-tang/page/2/) [home](https://md.research.elamur.ai/free/en/) / [authors](https://md.research.elamur.ai/free/en/authors/) / [Jie Tang](https://md.research.elamur.ai/free/en/authors/jie-tang/) / page 2 # Jie Tang · page 2 35 papers in our library · 2021–2026 · page 2 of 2 [The profile, affiliations and co-authors →](https://md.research.elamur.ai/free/en/authors/jie-tang/) ## Papers The most cited first (OpenAlex and our library). [← prev](https://md.research.elamur.ai/free/en/authors/jie-tang/) [1](https://md.research.elamur.ai/free/en/authors/jie-tang/) **2** 26. **[NaturalCodeBench: Examining Coding Performance Mismatch on HumanEval and Natural User Queries](https://md.research.elamur.ai/free/en/papers/zhang-2024-naturalcodebench-examining-coding-performance-mismatch/)** *(2024 · 17 citations)* Popular code-generation benchmarks such as HumanEval, MBPP and DS-1000 mostly test short, well-specified algorithmic or data-science exercises. 27. **[Scaling Speech-Text Pre-training with Synthetic Interleaved Data](https://md.research.elamur.ai/free/en/papers/zeng-2024-scaling-speech-text-pre-training/)** *(2024 · 14 citations)* Speech language models (SpeechLMs) have far less training material than text LLMs. FineWeb offers about 15T text tokens, while a large unsupervised speech set like VoxPopuli gives only around 36B tokens. Parallel speech-text data is also limited. The authors work around this by creating speech-text interleaved… 28. **[DataSciBench: An LLM Agent Benchmark for Data Science](https://md.research.elamur.ai/free/en/papers/zhang-2026-datascibench-llm-agent-benchmark-data/)** *(2026 · 13 citations)* Most benchmarks for LLMs in data science test isolated tasks with simple metrics such as accuracy and with ground truth that is easy to obtain. 29. **[MathGLM-Vision: Solving Mathematical Problems with Multi-Modal Large Language Model](https://md.research.elamur.ai/free/en/papers/yang-2024-mathglm-vision-solving-mathematical-problems/)** *(2024 · 10 citations)* Most math-focused multi-modal LLMs (MLLMs) are trained almost entirely on geometry, using small public sets such as GeoQA and Geometry3K whose answers rarely show the reasoning. 30. **[Glyph: Scaling Context Windows via Visual-Text Compression](https://md.research.elamur.ai/free/en/papers/cheng-2026-glyph-scaling-context-windows-visual/)** *(2026 · 6 citations)* Long-context LLMs usually scale by handling more text tokens, which raises compute and memory costs. Glyph takes a different route. Instead of enlarging the token window, it packs more source text into each token: documents are rendered as images and fed to a vision-language model. Built on GLM-4.1V-9B-Base, the… 31. **[Beyond Literal Mapping: Benchmarking and Improving Non-Literal Translation Evaluation](https://md.research.elamur.ai/free/en/papers/tian-2026-beyond-literal-mapping-benchmarking-improving/)** *(2026 · 5 citations)* LLM-based machine translation is increasingly used for difficult content such as social media posts, culturally loaded idioms, poetry and literature. 32. **[Data Efficient RLVR via Off-Policy Influence Guidance](https://md.research.elamur.ai/free/en/papers/zhu-2026-data-efficient-rlvr-off-policy/)** *(2026)* Choosing training data for reinforcement learning with verifiable rewards (RLVR) usually relies on heuristics such as difficulty or uncertainty. 33. **[KARL: Reinforcement Learning for LLM Agents on Multi-Turn Knowledge-Intensive Agentic Tasks](https://md.research.elamur.ai/free/en/papers/sun-2026-karl-reinforcement-learning-llm-agents/)** *(2026)* LLM agents often handle knowledge-heavy tasks poorly because they consume external information passively. Retrieval is either fixed in advance or tuned only for single-turn question answering. KARL (Knowledge-Augmented Reinforcement Learning) addresses this for multi-turn work over structured sources such as… 34. **[MTP-RL: Acceleration of Reinforcement Learning Rollouts with Policy-Aligned Multi-Token Prediction](https://md.research.elamur.ai/free/en/papers/wang-2026-mtp-rl-acceleration-reinforcement-learning/)** *(2026)* Rollout generation consumes most of the compute in reinforcement-learning fine-tuning of LLMs, especially when models produce long reasoning chains. 35. **[PlotGen-Bench: Evaluating VLMs on Generating Visualization Code from Diverse Plots across Multiple Libraries](https://md.research.elamur.ai/free/en/papers/zhao-2026-plotgen-bench-evaluating-vlms-generating/)** *(2026)* Vision-language models can already read and reason about charts, but their ability to write runnable code that recreates a given plot has been tested only narrowly. [← prev](https://md.research.elamur.ai/free/en/authors/jie-tang/) [1](https://md.research.elamur.ai/free/en/authors/jie-tang/) **2**