XHotpotQA is a new benchmark for cross‑lingual knowledge composition in multi‑hop question answering. It presents each instance as an evidence‑dependency graph with explicit language assignments for the question, bridge evidence, answer‑bearing evidence, and distractors, and includes 15,661 training and 7,405 validation examples with sentence‑level support supervision. The dataset reveals significant performance drops when evidence spans language boundaries, providing a diagnostic tool for systems that must integrate evidence across languages.
arXiv:2501. 06286v2 Announce Type: replace-cross Abstract: Multi-hop question answering requires a system to identify and integrate evidence distributed across documents, yet large language models remain vulnerable to irrelevant context.
arXiv:2609.37469v1 Announce Type: cross Abstract: Retrieval-augmented generation (RAG) grounds large language models in external sources, but retrieved passages often name the right entities without...
The paper introduces a trace-level divergence framework for evaluating multilingual large language models, breaking down agreement into structural similarity, order consistency, and holistic similarity. Experiments across three reasoning benchmarks, seven languages, and six leading LLMs show that conventional holistic similarity metrics overstate agreement, with up to 67.3% of top-scoring examples lacking structural alignment. Human judgments align more closely with structural similarity, highlighting language-specific divergence and the need for trace-level evaluation in multilingual deployments.
arXiv:2607. 19243v1 Announce Type: cross Abstract: Although Large Language Models (LLMs) demonstrate remarkable multilingual fluency, their internal knowledge representations remain disproportionately biased toward high-resource languages.
arXiv:2610.26951v1 Announce Type: new Abstract: Retrieval-augmented generation (RAG) is the dominant paradigm for grounding large language models in external corpora, but its similarity-driven, chunk...
arXiv:2607. 10562v1 Announce Type: new Abstract: Evaluating the multi-hop reasoning capabilities of large language models remains a significant challenge.
arXiv:2608. 10444v1 Announce Type: cross Abstract: Large language models (LLMs) have made substantial progress on reasoning tasks that require increasingly long and complex inferential chains.
arXiv:2608. 03292v1 Announce Type: new Abstract: Long Document Visual Question Answering (LongDocVQA) requires Multimodal Large Language Models (MLLMs) to locate, integrate, and reason over heterogeneous document elements distributed across multiple pages.
The paper introduces RAG‑Stress, a diagnostic protocol that tests how retrieval‑augmented generation models handle misleading evidence. By fixing the question and reference answer, editing a single assertion to support an incorrect answer, and varying source‑priority policies and evidence positions, the study measures misleading rate (MR) and clean accuracy across fifteen systems on multiple QA datasets. Results show that prioritizing documents increases MR by 10.9–13.5 percentage points, with the highest MR when the misleading span appears at the end of the evidence.
The study investigates the performance gap between native-language reasoning and English-pivoted reasoning in large language models. By creating extensive multilingual reasoning datasets and fine‑tuning specialists on Qwen/Qwen3-8B-Base, the authors find that the native reasoning gap is much smaller (1.9–3.5%) than previously reported. They analyze weight‑space changes, discover a language‑agnostic reasoning core in the middle layers, and propose a Layer Swap technique that transfers these mid‑layer updates from an English specialist to native specialists, effectively closing most of the gap while maintaining native chain‑of‑thought output.
The paper introduces TRACE, a fine‑tuning framework for Retrieval‑Augmented Generation (RAG) that addresses conflicts between retrieved knowledge and a model’s internal knowledge. TRACE uses multi‑agent debate traces to identify correct and incorrect candidates and answer‑shift patterns, providing fine‑grained supervision for reliable knowledge‑source selection. It also incorporates an answer‑completeness regularization mechanism to prevent empty, overly short, or prematurely terminated responses, thereby improving robustness against misleading retrieved content and enhancing answer quality.
arXiv:2608. 15535v1 Announce Type: cross Abstract: We present L3Cube-IndicQuest v2, a large-scale gold-standard multilingual question-answering benchmark for evaluating the India-specific factual knowledge of Large Language Models (LLMs).