Papers with semantic drift
Extracting Common Inference Patterns from Semi-Structured Explanations (D19-60)
Copied to clipboard
| Challenge: | Multi-hop inference suffers from semantic drift, or the tendency for chains of reasoning to "drift"' to unrelated topics. |
| Approach: | They propose to extract large high-confidence multi-hop inference patterns from a corpus of explanations by abstracting large-scale structure from logical sentences. |
| Outcome: | The proposed method extracts large high-confidence multi-hop inference patterns from a “matter” subset of elementary science exam questions. |
RiskLab: A Controlled Toolkit for Probing Emergent Risks in LLM-Based Multi-Agent Systems (2026.acl-demo)
Copied to clipboard
Yu Jiang, Wenjie Wang, Yue Huang, Yanbo Wang, Zhenhong Zhou, Xiuying Chen, Yang Liu, Pin-Yu Chen, Wei Wang, Xiangliang Zhang
| Challenge: | Recent advances in large language model (LLM) agents have accelerated deployment of multi-agent systems for complex tasks. |
| Approach: | They propose an open-source toolkit for instantiating, probing, and measuring emergent risks in LLM-based multi-agent systems under controlled conditions. |
| Outcome: | The proposed toolkit is based on a structured topology–environment–protocol–agent–task quintuple enabling reproducible studies of how communication structure, coordination mechanisms, and incentives shape system-level risks. |
Not All Directions Matter: Towards Structured and Task-Aware Low-Rank Model Adaptation (2026.acl-long)
Copied to clipboard
Xi Xiao, Chenrui Ma, Yunbei Zhang, Chen Liu, Zhuxuanzi Wang, Yanshu Li, Lin Zhao, Guosheng Hu, Tianyang Wang, Hao Xu
| Challenge: | Low-Rank Adaptation (LoRA) is a key parameter-efficient fine-tuning method . however, its effectiveness is hampered by semantic drift and structural incoherence . |
| Approach: | They propose a low-rank Adaptation framework that tackles semantic drift and structural incoherence by pruning task-irrelevant directions. |
| Outcome: | Experiments on large language models, vision models, and vision models show that the proposed framework outperforms LoRA and advanced dynamic rank allocation and sparsity-based methods. |
Bootstrapping Polar-Opposite Emotion Dimensions from Online Reviews (L18-1)
Copied to clipboard
| Challenge: | Existing bootstrapping methods for learning lexicons from unannotated online texts have important drawbacks. |
| Approach: | They propose a bootstrapping approach that softly labels unlabeled terms for polar-opposite emotion dimension values from the Ortony/Clore/Collins model of emotions. |
| Outcome: | The proposed approach achieves considerably better performance than several baselines. |
Stand on The Shoulders of Giants: Building JailExpert from Previous Attack Experience (2025.emnlp-main)
Copied to clipboard
Xi Wang, Songlei Jian, Shasha Li, Xiaopeng Li, Bin Ji, Ma Jun, Xiaodong Liu, Jing Wang, Jianfeng Zhang, Jie Yu, Feilong Bao, null Wangbaosheng
| Challenge: | Existing methods to generate human-aligned content with a “jailbreak prompt” are inefficient and repetitive, causing inefficiency and a lack of experience. |
| Approach: | They propose a framework that integrates past attack experiences to aid current jailbreak attempts. |
| Outcome: | The proposed framework improves both attack effectiveness and efficiency compared to the current black-box jailbreak method. |
Breaking the Static Graph: Context-Aware Traversal for Graph-Based RAG (2026.findings-acl)
Copied to clipboard
| Challenge: | Recent advances in RAG focus on capturing multi-hop dependencies, but static Graphs fail to retrieve complete evidence chain. |
| Approach: | They propose a structure-aware approach to capture multi-hop dependencies using Knowledge Graphs and Personalized PageRank to capture semantic drift. |
| Outcome: | Experiments show that CatRAG outperforms state-of-the-art approaches . the proposed approach achieves substantial improvements in reasoning completeness . |
MiniELM: A Lightweight and Adaptive Query Rewriting Framework for E-Commerce Search Optimization (2025.findings-acl)
Copied to clipboard
| Challenge: | Existing methods for rewriting query terms struggle with natural language understanding . generative methods face high inference latency and cost in offline settings . |
| Approach: | They propose a hybrid pipeline for rewriting query queries using offline knowledge distillation and online reinforcement learning. |
| Outcome: | The proposed pipeline improves query relevance, diversity, adaptability and cost-effective evaluation without manual annotations on Amazon ESCI dataset. |
Context as a Tool: Context Management for Long-Horizon SWE-Agents (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing large language models rely on append-only context maintenance or passively triggered compression heuristics, leading to context explosion, semantic drift, and degraded reasoning in long-running interactions. |
| Approach: | They propose a new context management paradigm that elevates context maintenance to a callable tool . they propose 'cat' framework that injects context-management actions into complete interaction trajectories . |
| Outcome: | The proposed model outperforms ReAct-based agents and static compression baselines on SWE-Verified tests. |
Sem-DPO: Mitigating Semantic Inconsistency in Preference Optimization for Prompt Engineering (2026.findings-acl)
Copied to clipboard
Anas Mohamed, Azal Ahmad Khan, Xinran Wang, Ahmad Faraz Khan, Shuwen Ge, Saman Bahzad Khan, Ayaan Ahmad, Ali Anwar
| Challenge: | Direct Preference Optimization (DPO) is an off-policy alternative to RL for automatic prompt engineering, but its token-level regularization leaves semantic inconsistency unchecked as prompts that win higher preference scores can still drift away from the user’s intended meaning. |
| Approach: | They propose a variant of Direct Preference Optimization that preserves semantic consistency while maintaining its simplicity and efficiency. |
| Outcome: | The proposed model outperforms state-of-the-art prompt optimization baselines and several DPO variants on three standard text-to-image prompt-optimization benchmarks and three language models. |
Mechanistic Interpretability of Text-to-Image Diffusion Models via Cross-Attention Interventions (2026.findings-acl)
Copied to clipboard
| Challenge: | Text-to-image diffusion models generate high quality images through iterative denoising, but their internal mechanisms for grounding prompt semantics into visual structure remain unclear. |
| Approach: | They propose a mechanistic interpretability framework that probes how individual prompt tokens are represented and utilized during the denoising process. |
| Outcome: | The proposed framework enables module-wise and head-wise attribution of semantic changes across denoising timesteps. |
Large Language Models Threaten Language’s Epistemic and Communicative Foundations (2025.emnlp-main)
Copied to clipboard
| Challenge: | Large language models are reshaping the norms of human communication, sometimes decouping words from genuine human thought. |
| Approach: | They propose to model humans, LLMs, and texts in a provenance network . they propose to use epistemic doppelgängers to generate texts that are indis- tinguishable from human-authored texts . |
| Outcome: | The proposed models induce semantic drift, erode account-ability, and obfuscate intent and authorship. |
Recovered in Translation: Efficient Pipeline for Automated Translation of Benchmarks and Datasets (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing benchmarks suffer from semantic drift and context loss, which can lead to misleading performance metrics. |
| Approach: | They propose a fully automated framework to enable translation of large language models . they propose to use universal self-improvement and multi-round ranking methods to improve translation quality . |
| Outcome: | The proposed framework surpasses existing benchmarks in eight languages and improves translation quality across multilingual domains. |