--- title: "Bing Qin — papers, page 2 · Harbin Institute of Technology" description: "Bing Qin, Harbin Institute of Technology: 43 papers on Elamur Research (2014–2026), h-index 47 — affiliations, topics, co-authors and who…" canonical: "https://research.elamur.ai/free/en/authors/bing-qin/page/2/" --- [Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/authors/bing-qin/page/2/) [home](https://md.research.elamur.ai/free/en/) / [authors](https://md.research.elamur.ai/free/en/authors/) / [Bing Qin](https://md.research.elamur.ai/free/en/authors/bing-qin/) / page 2 # Bing Qin · page 2 43 papers in our library · 2014–2026 · page 2 of 2 [The profile, affiliations and co-authors →](https://md.research.elamur.ai/free/en/authors/bing-qin/) ## Papers The most cited first (OpenAlex and our library). [← prev](https://md.research.elamur.ai/free/en/authors/bing-qin/) [1](https://md.research.elamur.ai/free/en/authors/bing-qin/) **2** 26. **[MPO: Multilingual Safety Alignment via Reward Gap Optimization](https://md.research.elamur.ai/free/en/papers/zhao-2025-mpo-multilingual-safety-alignment-reward/)** *(2025 · 7 citations)* Safety-tuned LLMs refuse harmful requests reliably in English but much less so in lower-resource languages. 27. **[SSR: Utilizing Simplified Stance Reasoning Process for Robust Stance Detection](https://md.research.elamur.ai/free/en/papers/yuan-2022-ssr-utilizing-simplified-stance-reasoning/)** *(2022 · 6 citations)* Stance detection models trained on benchmarks such as SemEval2016 Task 6 often score well even when the target is ignored. 28. **[Culture-Aware Machine Translation in Large Language Models: Benchmarking and Investigation](https://md.research.elamur.ai/free/en/papers/yuan-2026-culture-aware-machine-translation-large/)** *(2026 · 2 citations)* LLMs translate general text well, but little is known about how they handle culturally loaded content. The authors build CanMT, a parallel benchmark drawn from classic novels and their professional translations. It covers 12 directions among English, Chinese, Japanese, Spanish and Russian. 29. **[Psychological Counseling Cannot Be Achieved Overnight: Automated Psychological Counseling Through Multi-Session Conversations](https://md.research.elamur.ai/free/en/papers/wang-2026-psychological-counseling-cannot-be-achieved/)** *(2026 · 2 citations)* Most LLM-based counseling systems treat therapy as a single conversation. Real therapy such as Cognitive Behavioral Therapy (CBT) unfolds over many visits, with goals revised as the client changes. This paper frames automated counseling as a longitudinal task made of three parts: memory extraction, goal planning and… 30. **[MAESTRO: Meta-learning Adaptive Estimation of Scalarization Trade-offs for Reward Optimization](https://md.research.elamur.ai/free/en/papers/zhao-2026-maestro-meta-learning-adaptive-estimation/)** *(2026 · 1 citation)* GRPO aligns LLMs well when answers can be checked automatically, as in math or code. Open-ended generation has no such check, and its goals often conflict, for example creativity versus factuality. Existing fixes either rely on costly LLM judges or combine cheap heuristic rewards (perplexity, entropy, format) using… 31. **[On Safety Risks in Experience-Driven Self-Evolving Agents](https://md.research.elamur.ai/free/en/papers/zhao-2026-safety-risks-experience-driven-self/)** *(2026 · 1 citation)* Experience-driven self-evolving LLM agents improve by distilling lessons from past interactions into an external memory and retrieving them for new tasks, while model weights stay frozen. 32. **[TEA-Bench: A Systematic Benchmarking of Tool-enhanced Emotional Support Dialogue Agent](https://md.research.elamur.ai/free/en/papers/sui-2026-tea-bench-systematic-benchmarking-tool/)** *(2026 · 1 citation)* Most emotional support conversation (ESC) systems and benchmarks only measure empathetic wording in text-only settings. 33. **[x1: Learning to Think Adaptively Across Languages and Cultures](https://md.research.elamur.ai/free/en/papers/ye-2026-x1-learning-think-adaptively-across/)** *(2026 · 1 citation)* Most reasoning LLMs think mainly in one dominant language, usually English, even when a question is posed in another language or depends on a specific culture. 34. **[Adaptive Backtracking for Privacy Protection in Large Language Models](https://md.research.elamur.ai/free/en/papers/yao-2026-adaptive-backtracking-privacy-protection-large/)** *(2026)* In retrieval-augmented generation, attackers can write prompts that make an LLM reveal private details from the documents it retrieves. 35. **[Consolidation or Adaptation? PRISM: Disentangling SFT and RL Data via Gradient Concentration](https://md.research.elamur.ai/free/en/papers/zhao-2026-consolidation-or-adaptation-prism-disentangling/)** *(2026)* The usual recipe for training LLM agents runs a supervised fine-tuning (SFT) stage first and a reinforcement learning (RL) stage after it. 36. **[Fine-Mem: Fine-Grained Feedback Alignment for Long-Horizon Memory Management](https://md.research.elamur.ai/free/en/papers/ma-2026-fine-mem-fine-grained-feedback/)** *(2026)* LLM agents working on long-horizon tasks need a memory system that turns a stream of incoming text chunks into a compact, usable store. 37. **[From Implicit Graph Encoding to Explicit Evidence: A Training-Free LLM Framework for Temporal Knowledge Graph Reasoning](https://md.research.elamur.ai/free/en/papers/tang-2026-implicit-graph-encoding-explicit-evidence/)** *(2026)* Forecasting future facts in Temporal Knowledge Graphs (TKGs) is difficult for parametric models. They degrade under distribution shift and cannot handle unseen entities or relations without retraining. Existing LLM-based alternatives have their own problems: some flatten graphs into long text and lose information… 38. **[Large Language Models Are Still Misled by Simple Bias Ensembles](https://md.research.elamur.ai/free/en/papers/sun-2026-large-language-models-are-still/)** *(2026)* Recent LLMs handle individual shortcut cues better than before. This paper argues that the real weakness appears when several such cues occur together in one input, which is common in real-world text. Most existing bias benchmarks inject only one bias per sample, and a few inject two. The authors therefore built… 39. **[MDC-Bench: A Multidisciplinary Causal Benchmark Based on Causal Structures for Evaluating Large Language Models](https://md.research.elamur.ai/free/en/papers/wang-2026-mdc-bench-multidisciplinary-causal-benchmark/)** *(2026)* Most existing causal benchmarks for LLMs use everyday commonsense scenarios with short, single-hop cause–effect links. 40. **[PARIF: Pushing the Pareto Frontier of Instruction Following and Reasoning with Curriculum Reinforcement Learning](https://md.research.elamur.ai/free/en/papers/mu-2026-parif-pushing-pareto-frontier-instruction/)** *(2026)* Large Reasoning Models handle complex problems well but often ignore explicit constraints in user instructions. 41. **[Question Tells You Where the Answer Is: Intention-aware Long-Context KV Cache Compression](https://md.research.elamur.ai/free/en/papers/zhao-2026-question-tells-you-where-answer/)** *(2026)* Long context windows make the key-value (KV) cache of large language models very expensive in GPU memory and decoding time. 42. **[WebAnchor: Anchoring Agent Planning to Stabilize Long-Horizon Web Reasoning](https://md.research.elamur.ai/free/en/papers/yu-2026-webanchor-anchoring-agent-planning-stabilize/)** *(2026)* LLM-based web research agents trained with reinforcement learning often lose strategic coherence over long multi-step searches. 43. **[When Correct Beliefs Collapse: Epistemic Resilience of LLMs under Clinical Pressure](https://md.research.elamur.ai/free/en/papers/xiao-2026-when-correct-beliefs-collapse-epistemic/)** *(2026)* Medical LLMs score highly on static exam-style benchmarks, but this work asks whether they hold onto a correct diagnosis when a user keeps pushing back over several turns. [← prev](https://md.research.elamur.ai/free/en/authors/bing-qin/) [1](https://md.research.elamur.ai/free/en/authors/bing-qin/) **2**