Existing financial-NLP benchmarks mostly evaluate prepared artifacts such as filings, tables, or extracted values. Real accounting begins earlier: source documents must be reconciled into cited journal entries, aggregated into a balance sheet, and checked for contradictions. We introduce FINBALANCE, a multi-document accounting reconciliation benchmark built from source-document bundles across eight industries, three period types, and five difficulty levels. Human-authored business scenarios, accounting policies, tax/FX treatments, document schemas, distractors, and inconsistency templates are composed by a deterministic generator whose ledger produces journal entries, balance sheets, and 23 inconsistency-code labels. On a 710-record evaluation split, six contemporary large language models (LLMs) reach at most 46% exact final-balance-sheet accuracy. For four models, balance sheets reconstructed from their journal entries score 26–41 percentage points higher than their own reported balance sheets. Models often recover numerically plausible entries but fail to bind them to supporting documents and aggregate them consistently. Citation-pressure prompting barely changes document-linking errors, while ledger-feedback ablations substantially improve reported balance sheets and expose inconsistency-detection trade-offs. Expert finance reviewers validate the benchmark design and labels.

As researchers increasingly adopt LLMs as writing assistants, generating high-quality research paper introductions remains both challenging and essential. In this work, we benchmark Scientific Introduction Generation (SciIG), evaluating how well LLMs can produce coherent introductions from titles, abstracts, and related works in the machine learning domain. Curating new datasets from NAACL 2025 and ICLR 2025 papers, we assess five state-of-the-art models, including both open-source (DeepSeek-v3, Gemma-3-12B, LLaMA-4 Maverick, MistralAI Small 3.1) and closed-source GPT-4o systems across multiple dimensions: lexical overlap, semantic similarity, content coverage, faithfulness, consistency, citation correctness, and narrative quality. Our comprehensive framework combines automated metrics with LLM-as-a-judge evaluations. Results demonstrate LLaMA-4 Maverick’s superior performance on most metrics, particularly in lexical overlap, content coverage, citation quality, and perplexity. Moreover, three-shot prompting outperforms fewer-shot approaches more than 50% of the time on formula-based automated metrics and performs consistently better on all human evaluation metrics. These findings provide practical insights into developing effective research writing assistants and set realistic expectations for LLM-assisted academic writing. To foster reproducibility and future research, we publicly release all code and datasets.[Project page, code, and datasets: <https://kgarg8.github.io/emnlp26-sciig/>]

The task of Stance Detection involves discerning the stance expressed in a text towards a specific subject or target. Prior works have relied on existing transformer models that lack the capability to prioritize targets effectively. Consequently, these models yield similar performance regardless of whether we utilize or disregard target information, undermining the task’s significance. To address this challenge, we introduce Stanceformer, a target-aware transformer model that incorporates enhanced attention towards the targets during both training and inference. Specifically, we design a Target Awareness matrix that increases the self-attention scores assigned to the targets. We demonstrate the efficacy of the Stanceformer with various BERT-based models, including state-of-the-art models and Large Language Models (LLMs), and evaluate its performance across three stance detection datasets, alongside a zero-shot dataset. Our approach Stanceformer not only provides superior performance but also generalizes even to other domains, such as Aspect-based Sentiment Analysis. We make the code publicly available.

Keyphrase generation is the task of summarizing the contents of any given article into a few salient phrases (or keyphrases). Existing works for the task mostly rely on large-scale annotated datasets, which are not easy to acquire. Very few works address the problem of keyphrase generation in low-resource settings, but they still rely on a lot of additional unlabeled data for pretraining and on automatic methods for pseudo-annotations. In this paper, we present data augmentation strategies specifically to address keyphrase generation in purely resource-constrained domains. We design techniques that use the full text of the articles to improve both present and absent keyphrase generation. We test our approach comprehensively on three datasets and show that the data augmentation strategies consistently improve the state-of-the-art performance. We release our source code at https://github.com/kgarg8/kpgen-lowres-data-aug.

Stance detection aims to detect the stance toward a corresponding target. Existing works use the assumption that the target is known in advance, which is often not the case in the wild. Given a text from social media platforms, the target information is often unknown due to implicit mentions in the source text and it is infeasible to have manual target annotations at a large scale. Therefore, in this paper, we propose a new task Target-Stance Extraction (TSE) that aims to extract the (target, stance) pair from the text. We benchmark the task by proposing a two-stage framework that first identifies the relevant target in the text and then detects the stance given the predicted target and text. Specifically, we first propose two different settings: Target Classification and Target Generation, to identify the potential target from a given text. Then we propose a multi-task approach that takes target prediction as the auxiliary task to detect the stance toward the predicted target. We evaluate the proposed framework on both in-target stance detection in which the test target is always seen in the training stage and zero-shot stance detection that needs to detect the stance for the targets that are unseen during the training phase. The new TSE task can facilitate future research in the field of stance detection.

Keyphrase generation aims at generating important phrases (keyphrases) that best describe a given document. In scholarly domains, current approaches have largely used only the title and abstract of the articles to generate keyphrases. In this paper, we comprehensively explore whether the integration of additional information from the full text of a given article or from semantically similar articles can be helpful for a neural keyphrase generation model or not. We discover that adding sentences from the full text, particularly in the form of the extractive summary of the article can significantly improve the generation of both types of keyphrases that are either present or absent from the text. Experimental results with three widely used models for keyphrase generation along with one of the latest transformer models suitable for longer documents, Longformer Encoder-Decoder (LED) validate the observation. We also present a new large-scale scholarly dataset FullTextKP for keyphrase generation. Unlike prior large-scale datasets, FullTextKP includes the full text of the articles along with the title and abstract. We release the source code at https://github.com/kgarg8/FullTextKP.