Summarization
-
Understanding Axes of Difficulty For Long Context Tasks Via PredicateLongBench
2026-07-09 09:28:06Siddhartha Jain, Ameya VelingkerAbstract
Large language models (LLMs) have demonstrated rapidly improving long-context capabilities, prompting a wave of benchmarks designed to evaluate them. However, existing long-context evaluations - from Needle-in-a-Haystack (NIAH) tests to more recent multi-hop reasoning and summarization tasks - predominantly measure average-case performance, and many are either saturated or lack robustness. Notably absent is a systematic way to probe how models perform as we scale up the difficulty of tasks along various axes. We address this gap by proposing PredicateLongBench, a benchmark that stress-tests long-context reasoning by asking models to identify the longest contiguous subsequence of words in a long input that satisfies given predicates/constraints (e.g., lexicographic ordering), drawn from a broader predicate class. The central innovation of our benchmark is the identification and systematic exploration of multiple different axes of difficulty which test multiple aspects of long context understanding. We provide two complementary generation pipelines - a fully synthetic setup using random word-like strings, and a real-world setup that samples words from natural documents while preserving their distributional properties. We find that frontier models struggle to perform well as we scale up the difficulty of tasks along our axes, demonstrating the utility of our benchmark in understanding the limitations of current long-context capabilities. Furthermore, the tasks in PredicateLongBench, though challenging, are conceptually simple and do not require LLM-based generations or judges.
Abstract (translated)
URL
https://arxiv.org/abs/2607.08284
PDF
-
Out of Sight: Compression-Aware Content Protection against Agentic Crawlers
Abstract
The rise of LLM-based agents with reasoning, summarization, and memory capabilities has created a new threat surface for online content that conventional defenses fail to address. Existing defenses like access controls can be circumvented by agents mimicking ordinary browsers, and injection-based defenses often degrade human readability. In this paper, we revisit the agent pipeline and identify context compression, which agents routinely invoke to fit context budgets, as a critical yet overlooked defense layer. We propose CAPE, a framework that protects high-value textual content by injecting invisible perturbations without changing its human-visible surface form, thereby inducing severe information loss during agent compression. CAPE extracts disruptive seed perturbations from an accessible surrogate compressor, then adapts them to query-only target compressors through prior-guided evolution and preference-calibrated candidate prioritization, achieving effective protection under a low query budget. Experiments on three content types and four compression settings show that CAPE improves information loss by up to 75.8% over the strongest baseline while keeping protected content visually indistinguishable from originals. CAPE also transfers to real-world settings, including the LangGraph agent workflow and GitHub Copilot, highlighting its generality and practical value. This paper aims to reveal context compression as a new defense layer, promoting content protection research in the agent era.
Abstract (translated)
URL
https://arxiv.org/abs/2607.08180
PDF
-
RWGBench: Evaluating Scholarly Positioning in Related Work Generation
2026-07-08 14:41:34Anzhe Xie, Weihang Su, Jiaxin Mao, Yiqun Liu, Shaoping Ma, Qingyao AiAbstract
Large language models have shown strong fluency in scientific writing, yet the evaluation of related work generation (RWG) remains limited. Existing RWG evaluations largely inherit summarization-oriented metrics, using lexical or semantic similarity to reference sections as proxies for quality. However, related work writing is fundamentally a citation-level scholarly positioning task: it requires selecting, organizing, and framing prior work to clarify how a target paper relates to, differs from, and contributes beyond existing this http URL a result, models may generate coherent and semantically-relevant text while exhibiting academically critical failures, such as inappropriate citation selection or misplaced references, that conventional metrics do not this http URL this end, we introduce \textbf{RWGBench}, a benchmark that evaluates RWG from the perspective of citation decision-making rather than text similarity. RWGBench is constructed from a large-scale collection of 40,108 computer science papers and a retrieval corpus of 1.09 million documents, with a carefully curated test set comprising 100 papers and their corresponding published related work this http URL propose a multi-dimensional evaluation framework that assesses citation selection, contextual appropriateness, organization, and discourse this http URL reveal systematic limitations in current systems that are obscured by standard evaluations, while Oracle studies further disentangle retrieval-level and generation-level bottlenecks. Human evaluation further shows that our citation-centric metrics align substantially better with expert judgment than surface-level text metrics. RWGBench offers a citation-centric testbed for developing and evaluating related work generation systems that are better aligned with scholarly writing practices.
Abstract (translated)
URL
https://arxiv.org/abs/2606.24894
PDF
-
DepthWeave-KV: Token-Adaptive Cross-Layer Residual Factorization for Long-Context KV Cache Compression
2026-07-07 17:29:01Anna Cordoba, Adam Puente Tercero, Nerea Angulo Hijo, Mar Linares Tercero, Julia Barrientos, Ainhoa Miranda, Jesus OliveraAbstract
Long-context language model inference is increasingly limited by the memory bandwidth and capacity required to store key-value caches, yet existing compression methods often apply uniform budgets across layers or tokens and degrade retrieval when lexical cues and semantic states require different preservation. We introduce DepthWeave-KV, a token-adaptive cache compression method that factorizes key and value states across neighboring transformer layers using shared low-rank channel bases while retaining lightweight token-specific residuals where attention behavior is sensitive. DepthWeave-KV combines cross-depth residual factorization with a token-conditional depth router that allocates higher reconstruction rank to instruction-bearing and retrieval-critical tokens, and uses calibration-free online error tracking from attention-output probes to adapt compression during generation without retraining the base model. A fused CUDA implementation jointly performs basis lookup, residual dequantization, and attention projection to reduce decode-time memory traffic. Across LongBench, Needle-in-a-Haystack, L-Eval, and long-form QA and summarization benchmarks, DepthWeave-KV achieves near-full-cache task quality with substantially lower memory use, improving average score and retrieval accuracy over prior compressed caches while reaching 8.3x KV memory reduction and 72.8 tokens per second at 64K context.
Abstract (translated)
URL
https://arxiv.org/abs/2607.06523
PDF
-
FreqDepthKV: Frequency-Guided Depth Sharing for Robust KV Cache Compression in Long-Context LLM Inference
2026-07-07 17:26:28Anna C\'ordoba, Adam Puente Tercero, Nerea Angulo Hijo, Mar Linares Tercero, Julia Barrientos, Ainhoa Miranda, Jes\'us OliveraAbstract
Long-context LLM inference is increasingly limited by the memory and bandwidth cost of KV caches, yet aggressive compression can remove the layer-specific evidence needed for retrieval and multi-step reasoning. We introduce FreqDepthKV, an inference-time cache compression method that factorizes adjacent-layer KV states into shared low-frequency depth components and sparse high-frequency residuals. A lightweight online probe assigns attention heads to shared-depth, residual-depth, or exact cache modes according to their contribution to reconstruction-sensitive attention logits, allowing the compression policy to adapt to prompt structure without retraining. Across long-context question answering, needle retrieval, summarization, and code generation benchmarks, FreqDepthKV preserves task accuracy under substantially smaller cache budgets. With a 32k-token prefill window, FreqDepthKV reaches 58.3 Exact Match, 63.0 F1, 32.5 ROUGE-L, and 48.1 pass@1, closely matching full KV while outperforming prior compressed-cache methods. It also improves decoding throughput to 70.4 tokens/s, reduces TTFT to 2.06 seconds, and lowers peak KV memory to 6.2 GB, achieving a 3.9x effective compression ratio.
Abstract (translated)
URL
https://arxiv.org/abs/2607.06519
PDF
-
SelfMem: Self-Optimizing Memory for AI Agents
2026-07-04 06:27:19Shu Yang, Junchao Wu, Derek F. Wong, Di WangAbstract
While current AI agents support increasingly long context windows, tool use, and skill execution for long-horizon tasks, they still require memory systems to effectively leverage historical experience. Existing memory frameworks typically rely on fixed storage, retrieval, and summarization mechanisms, which can be rigid across different tasks and often require manual tuning. To address this limitation, we propose SelfMem, a self-optimizing memory framework. Inspired by prior work on self-improving AI, we follow the principle of "teaching an agent to fish rather than giving it a fish." Instead of forcing the model to follow a predefined memory strategy or format, SelfMem provides an environment with memory tools and feedback signals that allow the agent to explore, evaluate, and refine its own memory strategy. Our results show that SelfMem consistently outperforms retrieval, compression, and agent-memory baselines on BEAM across conversation scales from 100K to 1M tokens. Compared with the strongest baseline, SelfMem improves the official score by 48.7%, 40.8%, and 41.9% at 100K, 500K, and 1M, respectively. Further question-type analysis shows broad robustness across diverse memory demands, and our optimization study shows that model-guided strategy refinement further improves performance.
Abstract (translated)
URL
https://arxiv.org/abs/2607.03726
PDF
-
RADIO1D: Elastic Representations for Condensed Vision Modeling
2026-07-03 22:58:54Greg Heinrich, Mike Ranzinger, Collin McCarthy, Natan Bagrov, Eugene Khvedchenya, Bryan Catanzaro, Jan Kautz, Andrew Tao, Pavlo MolchanovAbstract
This paper challenges the assumption that vision-language models (VLMs) require fixed patch-based 2D vision features. Analyzing fine-tuned vision encoders, we find that representations become increasingly abstract and less spatially coherent during VLM training. Notably, models trained with image-text alignment (such as SigLIP2) develop a small number of specialized tokens that effectively summarize global image content. Building on this, we introduce RADIO1D, which compresses images into a compact, variable-length 1D token sequence using multi-teacher knowledge distillation and an autoencoder design. The resulting representations exhibit strong hierarchical summarization, enabling accurate scene understanding - even with a single token - and support improved composition-aware image retrieval. In VLMs, RADIO1D provides flexible accuracy-efficiency tradeoffs through adjustable token counts, delivering competitive performance on diverse multimodal benchmarks with lower computational overhead and better accuracy.
Abstract (translated)
URL
https://arxiv.org/abs/2607.03624
PDF
-
psytechlab at CLPsych 2026: Utilising Natural Language Processing methods and Large Language Models for Social Media Text Analysis
2026-07-03 06:37:57Igor Buyanov, Nafisa Valieva, Ekaterina MazurinaAbstract
Social media posts are a rich and valuable source of data for analyzing mental health states and users' well-being using automated analysis tools. In this work, we demonstrate how we used a range of Natural Language Processing (NLP) methods, including Long Short-Term Memory (LSTM), BERT-based models, and Large Language Models (LLMs), for self-state and well-being analysis and summarization during the CLPsych Shared Task 2026. Our approach achieved one of the top Consistency and Contradiction scores for the summarization task and also middle-level results for the other tasks. By testing and developing such mental health-state estimation systems, we contributed to improving mental health support systems. We make our code available this https URL.
Abstract (translated)
URL
https://arxiv.org/abs/2607.03003
PDF
-
An Exploratory Study on LLM-Generated Code and Comments in Code Repositories
2026-07-02 08:25:15Yongyi Ji, Jiaji Wang, Yi Zhou, Fuxiang Chen, Hongji YangAbstract
The use of LLMs in software development has become increasingly widespread on tasks such as code generation and summarization. Reports from large technology companies showed that around 20% to 30% of their code are generated by LLMs. However, there remains skepticism about the practical usage of LLM-generated code and comments, such as concerns on more time for debugging the generated code and the unnaturalness of the generated comments. In this paper, we study the code and comments detected as likely to be generated by LLMs and their characteristics, the differences between company- and community-maintained repositories, and how likely bugs are associated with LLM-generated code. We conduct extensive experiments on active company- and community-maintained repositories from 2021 to 2025 using various tools and techniques that detect code and comments generated by LLMs. Based on our detector-based proxy analysis, the results suggest that code detected as likely to be generated by LLMs decreased over time and appeared frequently in test cases, while that of comments remains relatively stable. Proxy results further suggest that code detected as likely to be generated by LLMs shows substantial intra-repository code clones, whereas comments exhibit a relatively low proportion of grammatically correct sentences. In addition, the company-maintained repositories show a higher percentage of code and comments detected as likely to be generated by LLMs, and only a small percentage of the human-labelled bugs are detected as being likely associated with LLM-generated code.
Abstract (translated)
URL
https://arxiv.org/abs/2607.01867
PDF
-
On the Limits of Steering Vectors for Preference-Aligned Generation
2026-07-02 07:18:36Melanie Subbiah, Zara Hall, Kathleen McKeownAbstract
Steering vectors have emerged as a promising approach to controlled text generation, offering interpretable, training-free mechanisms for shaping model outputs. However, their practical generality remains poorly understood. We study the limits of steering vector generalization along three dimensions: trait expressibility, task transfer, and multi-trait composition. Using the PLUME writing personalization benchmark, we extract steering vectors for a range of preferences and evaluate them on summarization and email-writing tasks across two open-source models (Qwen2.5-7B-Instruct and Llama3.1-8B-Instruct). We find that steering effectiveness varies substantially across traits. We further show that steering effectiveness can degrade when vectors extracted from positive and negative style examples are transferred to downstream writing personalization tasks. Finally, we compare common methods for composing multiple steering vectors and find that all methods suffer significant drops in trait expression as more vectors are added, with a tradeoff between coherence and expressibility that requires per-setting hyperparameter tuning. Taken together, our results suggest that steering vectors face meaningful limits as a general-purpose tool for preference alignment.
Abstract (translated)
URL
https://arxiv.org/abs/2607.01802
PDF
-
Agent4cs: A Multi-agent System for Code Summarization in Large Hierarchical Codebases
2026-07-01 19:41:38Yongjian Tang, Ezgi Sarikayak, Doruk Tuncel, Jie M. Zhang, Thomas RunklerAbstract
Understanding large, complex codebases, especially those with obfuscated structures and incomplete documentation, remains a significant challenge. Existing code summarization solutions often rely on a single language model or coding assistant like Claude Code, and treat source code as flat text, underutilizing the rich interdependencies and hierarchical information within a repository. To address these shortcomings, we propose Agent4cs - a multi-agent framework that summarizes large codebases in a bottom-up fashion, where a summarization agent focuses on producing robust summaries; a keyword-extraction agent proactively identifies critical information from subfolders; and a quality-assurance agent iteratively refines the outputs for readability, coherence, and completeness. Evaluated on 7 frontier models, Agent4cs improves semantic consistency across all folder levels by average 8% compared to two structured prompting baselines with code segments. Furthermore, extensive evaluation on real-world datasets demonstrates up to 38% gains in normalized keyword coverage rate over the same baselines.
Abstract (translated)
URL
https://arxiv.org/abs/2607.01425
PDF
-
Imprint: Online Memory Compression for Long-Horizon Egocentric QA
2026-07-01 09:45:30Kousik Das, Debaditya RoyAbstract
Long-horizon egocentric question answering involves answering about events that have occurred hours or days in the past. This requires memory representations that remain both retrieval-effective and scalable over days or weeks of recording. Existing long-horizon egocentric QA methods construct memory as hierarchical textual summaries of observations. While effective for reducing memory size, summarization optimizes for descriptive compression rather than retrieval: repeated interactions are absorbed into coarse textual descriptions instead of being preserved as explicit, recurring memory units, making long-horizon evidence aggregation difficult. We propose Imprint, an interaction-centric memory framework that formulates long-horizon egocentric memory as an online memory compression problem rather than summarization. Incoming observations are first represented as structured Interaction Records and continuously organized into recurring interaction patterns. Using human memory consolidation signals of recurrence, recency, and distinctiveness, Imprint selectively retains and compresses interactions into a compact retrieval-oriented memory. We evaluate Imprint on EgoLifeQA, a seven-day egocentric benchmark containing questions that require reasoning over interactions occurring hours to days before the query. With the same LLM, Imprint improves QA accuracy from 31.0% to 35.8%, increases evidence-grounded answers by $6\times$ compared with EgoRAG, reduces memory footprint by $2.3\times$, and decreases retrieval latency by $11.8\times$. These results demonstrate that memory compression provides a scalable and retrieval-effective foundation for long-horizon egocentric question answering.
Abstract (translated)
URL
https://arxiv.org/abs/2607.00696
PDF
-
Measuring & Mitigating Over-Alignment for LLMs in Multilingual Criminal Law Courts
2026-06-30 12:22:05Arthur Wuhrmann, Gaetan Stein, Daniel Brunner, Andrei KucharavyAbstract
While the wider applicability of LLMs in the legal field is currently debated due to their reliability and the gravity of any errors, narrow uses with well-understood and mitigated risks have emerged. Notably the Swiss Federal Supreme Court uses small on-premises models for tentative translations and short-passage summarization across the four official languages. However, such usage is challenging in the context of Criminal Law. Since rulings and cases employees work on routinely can contain detailed descriptions of violent and sexual offenses, their legitimate work is compromised by refusals and disclaimers due to the activation of model guardrails (over-alignment). To measure this phenomenon, we introduce TF-RefusalBench, a multilingual benchmark for criminal-law translation and summarization derived from public Swiss Supreme Court rulings. TF-RefusalBench contains 5,200 total prompts across French, German, Italian, and English, corresponding to common task prompts and passages likely to trigger refusal. We then use TF-RefusalBench to show that over-alignment is a multifaceted phenomenon, influenced by the model and the prompt and text languages being processed, and that its impact cannot be evaluated solely from an over-refusal perspective, given the disclaimer's impact on task faithfulness. Finally, we evaluate approaches to enable on-premises LLMs for Criminal Law Tasks, demonstrating that while prompting can be effective, abliteration (refusal directions ablation) eliminates refusal with minimal impact on task performance.
Abstract (translated)
URL
https://arxiv.org/abs/2606.23375
PDF
-
ACE: Pluggable Adaptive Context Elasticizer across Agents
2026-06-30 12:20:45Ning Liao, Zihao Long, Xiaoxing Wang, Xue Yang, Yaoming Wang, Ziyuan Zhuang, Xunliang Cai, Rongxiang Weng, Junchi YanAbstract
The increasing complexity of agentic tasks has led to rapidly growing trajectory lengths, which poses significant challenges for large language model (LLM) based agents with fixed context windows. Existing context management techniques, such as truncation and summarization, suffer from inherent inflexibility and irreversibility: once information is discarded or compressed, it cannot be recovered even when it becomes critically relevant in later decision steps. To address these limitations, we propose the Adaptive Context Elasticizer (ACE), a plug-and-play module that elastically orchestrates historical step information into the agent's context at each decision step. ACE maintains a lossless message maintenance layer that stores both raw messages and compressed abstractions for each historical step, while a context orchestration layer adaptively assigns each step an elastic type as raw, abstract, or drop, at every decision step based on the current task state. This reversible design ensures that the main LLM always receives a compact yet information-rich context. We adapt ACE to four diverse agent frameworks, including ReAct, DeepAgent, WebThinker, and MiroFlow, without training or architectural modifications. Experiments show that ACE consistently outperforms truncation and summarization baselines, and brings consistent performance gains across all four agent frameworks.
Abstract (translated)
URL
https://arxiv.org/abs/2606.31564
PDF
-
Coverage-Driven KV Cache Eviction for Efficient and Improved Inference of LLM
2026-06-28 19:04:47Shuvendu Roy, Mengyao Zhai, Hossein Hajimirsadeghi, Golnoosh SameiAbstract
Large language models (LLMs) excel at complex tasks like question answering and summarization, thanks to their ability to handle long-context inputs. However, deploying LLMs is costly, not only due to the high computational demands of quadratic complexity of self-attention and auto-regressive generation, but also because of the significant memory overhead required for storing the key-value (KV) cache during inference. To reduce the memory cost, existing KV-cache eviction strategies leverage the sparsity in attention to selectively store a subset of tokens. While reducing the memory footprint, such approaches show a considerable drop in performance, especially in tasks that require long-context reasoning. We identify that the drop in performance is linked to a reduction in the coverage of unique tokens. Additionally, we theoretically show that reduced coverage limits the mutual information between inputs and outputs, thereby impairing predictive accuracy. To this end, we introduce K-VEC, a novel coverage-aware KV-cache eviction strategy that prioritizes token coverage while evicting tokens in the cache. K-VEC introduces a cross-head and a cross-layer coverage module to enhance token retention across attention heads and model layers, mitigating performance degradation caused by low coverage. Evaluated on 16 LongBench subsets, K-VEC exhibit up to 10.35 points improvement over the existing methods under the same eviction rate and memory constraint. Comprehensive evaluations validate the effectiveness of our approach and demonstrate its potential for efficient LLM deployment in resource-constrained settings.
Abstract (translated)
URL
https://arxiv.org/abs/2606.29563
PDF
-
Customized Generative AI Agent for Transportation Engineering Practice: A Development and Continued Pre-training Guideline
2026-06-27 17:22:13Dianwei Chen (Terry), Yuan-Zheng Lei (Terry), Zifan Zhang (Terry), Yuchen Liu (Terry), Xianfeng (Terry), YangAbstract
Recent advancements in generative artificial intelligence (AI) and large language models (LLMs) have shown significant promise in automating complex reasoning, summarization, and question-answering tasks. However, the effectiveness of general-purpose LLMs in specialized engineering domains remains limited due to insufficient exposure to technical standards, engineering terminology, and domain-specific semantics. This study proposes a systematic approach to developing a customized generative AI agent for transportation engineering applications. A curated corpus of U.S. transportation manuals, design guidelines, and regulatory documents is used to conduct continued pretraining of six state-of-the-art LLMs through a unified low-rank adaptation (LoRA) framework. The training process is monitored to ensure convergence and model stability. Performance is evaluated using standard natural language processing metrics, including BLEU-4 and ROUGE, with Qwen2.5-7B and LLaMA-3.1-8B demonstrating the highest domain alignment and response quality. Results validate the effectiveness of LoRA-based adaptation in improving LLM performance on technical content interpretation and context-specific reasoning. This work contributes a reproducible development framework for constructing domain-specialized generative AI agents, supporting broader deployment in transportation research, design, planning, and policy analysis.
Abstract (translated)
URL
https://arxiv.org/abs/2606.29014
PDF
-
Governance Decay: How Context Compaction Silently Erases Safety Constraints in Long-Horizon LLM Agents
2026-06-27 08:35:20Shiyang ChenAbstract
Modern LLM agents increasingly rely on context compaction, summarization, or eviction to keep long-running sessions within a token budget. We show that this context-management layer is a safety-critical failure surface: in-context governance constraints that agents reliably obey while visible can be silently removed by compaction, causing the same agent to perform prohibited tool actions later in the session. We call this failure mode Governance Decay. We introduce ConstraintRot, a benchmark of long-horizon agent scenarios with deterministic tool-call grading, and measure compaction-induced violations across seven model families. Across 1,323 episodes, violation rises from 0% with the policy in full context to 30% after compaction, reaching 59% for some models; when the constraint survives the summary, violation remains 0%, but when it is dropped, violation reaches 38%. We further study a Compaction-Eviction Attack, in which adversarial in-context content biases the summarizer to omit a legitimate policy, and show that optimized injections defeat every evaluated model. Finally, we propose Constraint Pinning, a simple training-free mitigation that quarantines governance constraints from lossy compaction and restores violation to 0% in our benchmark. These results identify context management as a first-class governance surface for deployed LLM agents.
Abstract (translated)
URL
https://arxiv.org/abs/2606.22528
PDF
-
A Tree-of-Thoughts Inspired Hybrid Approach for Legal Case Judgement Summarization using LLMs
2026-06-26 12:46:27Aniket Deroy, Kripabandhu Ghosh, Saptarshi GhoshAbstract
In recent times, Large Language Models (LLMs) are increasingly being used for legal case judgement summarization. Most prior works have tried traditional extractive and abstractive summarization of case judgements. However, hybrid or extractive-abstractive techniques have not been explored much. In this work, we propose a novel tree-of-thoughts inspired extractive-abstractive summarization approach for legal judgement summarization. We conduct experiments using two popular LLMs, DeepSeek and LLama, and compare among extractive, abstractive and extractive-abstractive summarization. Our experiments show that the proposed extractive-abstractive prompt provides better summaries compared to other types of LLM prompts.
Abstract (translated)
URL
https://arxiv.org/abs/2606.28044
PDF
-
Ask, Don't Judge: Binary Questions for Interpretable LLM Evaluation and Self-Improvement
2026-06-25 16:14:50Sangwoo Cho, Kushal Chawla, Pengshan Cai, Zefang Liu, Chenyang Zhu, Shi-Xiong Zhang, Sambit SahuAbstract
Evaluating LLM outputs remains a major bottleneck in NLP: human evaluation is expensive and slow, lexical metrics correlate poorly with human judgments on open-ended generation, and holistic LLM judges often produce opaque scores that are hard to debug. We propose BINEVAL, a framework that decomposes evaluation criteria into atomic binary questions and aggregates the resulting verdicts into interpretable, multi-dimensional scores. Given a task prompt, a meta-prompt generates fine-grained evaluation questions, and an LLM answers them independently for each output, yielding transparent question-level feedback together with calibrated overall scores. This decomposition makes evaluation easier to inspect, easier to diagnose, and directly usable for prompt improvement. Across SummEval, Topical-Chat, and QAGS, BINEVAL matches or outperforms strong baselines including UniEval and G-Eval, with especially strong results on factual consistency benchmarks such as QAGS. Beyond competitive correlation with human judgments, BINEVAL better matches human score distributions and avoids the ceiling effects common in prior LLM judges, leading to better discrimination between borderline and clearly flawed outputs. We further show that the same question-level feedback supports iterative prompt optimization, improving evaluator prompts on summarization and generation prompts on IFBench under both self-update and cross-model update settings. Overall, BINEVAL provides a task-agnostic, training-free, and interpretable evaluation framework that combines strong empirical performance with practical diagnostic and optimization value.
Abstract (translated)
URL
https://arxiv.org/abs/2606.27226
PDF
-
Detect, Unlearn, Restore: Defending Text Summarization Models Against Data Poisoning
2026-06-24 17:12:42Poojitha Thota, Shirin NilizadehAbstract
Training-time data poisoning during fine-tuning poses a significant threat to large language models (LLMs) deployed for abstractive text summarization, where small task-specific datasets exert disproportionate influence on model behavior. In this setting, adversaries manipulate fine-tuning data to induce persistent summarization failures, such as biased or harmful summaries, while preserving standard evaluation metrics. We present a unified post-hoc defense framework for detecting and remediating fine-tuning-stage poisoning in summarization models across the machine learning supply chain. Our experiments show that in white-box settings, poisoned document-summary pairs exhibit abnormally high training influence, enabling detection via influence-function analysis with semantic consistency checks. In black-box settings, poisoned models display two to three times greater sensitivity to semantics-preserving perturbations, enabling behavioral auditing without training data access. Beyond existing poisoning formulations, we introduce novel attacks targeting factual distortion and representational bias, showing that poisoning alters summarization behavior without triggering conventional alarms. Across nine architectures and six benchmark datasets under adaptive attacks, our defenses achieve 85-92% detection precision, while gradient-ascent unlearning restores up to 96% of original behavior with minimal utility loss (less than 0.6% ROUGE degradation). These results indicate that fine-tuning-time poisoning leaves persistent structural artifacts, enabling practical detection and post-deployment recovery without full retraining.
Abstract (translated)
URL
https://arxiv.org/abs/2606.26036
PDF
