ResearchRubrics: A Benchmark of Prompts and Rubrics For Evaluating Deep Research Agentscs.AI updates on arXiv.org arXiv:2511.07685v1 Announce Type: new
Abstract: Deep Research (DR) is an emerging agent application that leverages large language models (LLMs) to address open-ended queries. It requires the integration of several capabilities, including multi-step reasoning, cross-document synthesis, and the generation of evidence-backed, long-form answers. Evaluating DR remains challenging because responses are lengthy and diverse, admit many valid solutions, and often depend on dynamic information sources. We introduce ResearchRubrics, a standardized benchmark for DR built with over 2,800+ hours of human labor that pairs realistic, domain-diverse prompts with 2,500+ expert-written, fine-grained rubrics to assess factual grounding, reasoning soundness, and clarity. We also propose a new complexity framework for categorizing DR tasks along three axes: conceptual breadth, logical nesting, and exploration. In addition, we develop human and model-based evaluation protocols that measure rubric adherence for DR agents. We evaluate several state-of-the-art DR systems and find that even leading agents like Gemini’s DR and OpenAI’s DR achieve under 68% average compliance with our rubrics, primarily due to missed implicit context and inadequate reasoning about retrieved information. Our results highlight the need for robust, scalable assessment of deep research capabilities, to which end we release ResearchRubrics(including all prompts, rubrics, and evaluation code) to facilitate progress toward well-justified research assistants.
arXiv:2511.07685v1 Announce Type: new
Abstract: Deep Research (DR) is an emerging agent application that leverages large language models (LLMs) to address open-ended queries. It requires the integration of several capabilities, including multi-step reasoning, cross-document synthesis, and the generation of evidence-backed, long-form answers. Evaluating DR remains challenging because responses are lengthy and diverse, admit many valid solutions, and often depend on dynamic information sources. We introduce ResearchRubrics, a standardized benchmark for DR built with over 2,800+ hours of human labor that pairs realistic, domain-diverse prompts with 2,500+ expert-written, fine-grained rubrics to assess factual grounding, reasoning soundness, and clarity. We also propose a new complexity framework for categorizing DR tasks along three axes: conceptual breadth, logical nesting, and exploration. In addition, we develop human and model-based evaluation protocols that measure rubric adherence for DR agents. We evaluate several state-of-the-art DR systems and find that even leading agents like Gemini’s DR and OpenAI’s DR achieve under 68% average compliance with our rubrics, primarily due to missed implicit context and inadequate reasoning about retrieved information. Our results highlight the need for robust, scalable assessment of deep research capabilities, to which end we release ResearchRubrics(including all prompts, rubrics, and evaluation code) to facilitate progress toward well-justified research assistants. Read More
Text-based Aerial-Ground Person Retrievalcs.AI updates on arXiv.org arXiv:2511.08369v1 Announce Type: cross
Abstract: This work introduces Text-based Aerial-Ground Person Retrieval (TAG-PR), which aims to retrieve person images from heterogeneous aerial and ground views with textual descriptions. Unlike traditional Text-based Person Retrieval (T-PR), which focuses solely on ground-view images, TAG-PR introduces greater practical significance and presents unique challenges due to the large viewpoint discrepancy across images. To support this task, we contribute: (1) TAG-PEDES dataset, constructed from public benchmarks with automatically generated textual descriptions, enhanced by a diversified text generation paradigm to ensure robustness under view heterogeneity; and (2) TAG-CLIP, a novel retrieval framework that addresses view heterogeneity through a hierarchically-routed mixture of experts module to learn view-specific and view-agnostic features and a viewpoint decoupling strategy to decouple view-specific features for better cross-modal alignment. We evaluate the effectiveness of TAG-CLIP on both the proposed TAG-PEDES dataset and existing T-PR benchmarks. The dataset and code are available at https://github.com/Flame-Chasers/TAG-PR.
arXiv:2511.08369v1 Announce Type: cross
Abstract: This work introduces Text-based Aerial-Ground Person Retrieval (TAG-PR), which aims to retrieve person images from heterogeneous aerial and ground views with textual descriptions. Unlike traditional Text-based Person Retrieval (T-PR), which focuses solely on ground-view images, TAG-PR introduces greater practical significance and presents unique challenges due to the large viewpoint discrepancy across images. To support this task, we contribute: (1) TAG-PEDES dataset, constructed from public benchmarks with automatically generated textual descriptions, enhanced by a diversified text generation paradigm to ensure robustness under view heterogeneity; and (2) TAG-CLIP, a novel retrieval framework that addresses view heterogeneity through a hierarchically-routed mixture of experts module to learn view-specific and view-agnostic features and a viewpoint decoupling strategy to decouple view-specific features for better cross-modal alignment. We evaluate the effectiveness of TAG-CLIP on both the proposed TAG-PEDES dataset and existing T-PR benchmarks. The dataset and code are available at https://github.com/Flame-Chasers/TAG-PR. Read More
A Realistic Evaluation of Cross-Frequency Transfer Learning and Foundation Forecasting Modelscs.AI updates on arXiv.org arXiv:2509.19465v2 Announce Type: replace-cross
Abstract: Cross-frequency transfer learning (CFTL) has emerged as a popular framework for curating large-scale time series datasets to pre-train foundation forecasting models (FFMs). Although CFTL has shown promise, current benchmarking practices fall short of accurately assessing its performance. This shortcoming stems from many factors: an over-reliance on small-scale evaluation datasets; inadequate treatment of sample size when computing summary statistics; reporting of suboptimal statistical models; and failing to account for non-negligible risks of overlap between pre-training and test datasets. To address these limitations, we introduce a unified reimplementation of widely-adopted neural forecasting networks, adapting them for the CFTL setup; we pre-train only on proprietary and synthetic data, being careful to prevent test leakage; and we evaluate on 15 large, diverse public forecast competition datasets. Our empirical analysis reveals that statistical models’ accuracy is frequently underreported. Notably, we confirm that statistical models and their ensembles consistently outperform existing FFMs by more than 8.2% in sCRPS, and by more than 20% MASE, across datasets. However, we also find that synthetic dataset pre-training does improve the accuracy of a FFM by 7% percent.
arXiv:2509.19465v2 Announce Type: replace-cross
Abstract: Cross-frequency transfer learning (CFTL) has emerged as a popular framework for curating large-scale time series datasets to pre-train foundation forecasting models (FFMs). Although CFTL has shown promise, current benchmarking practices fall short of accurately assessing its performance. This shortcoming stems from many factors: an over-reliance on small-scale evaluation datasets; inadequate treatment of sample size when computing summary statistics; reporting of suboptimal statistical models; and failing to account for non-negligible risks of overlap between pre-training and test datasets. To address these limitations, we introduce a unified reimplementation of widely-adopted neural forecasting networks, adapting them for the CFTL setup; we pre-train only on proprietary and synthetic data, being careful to prevent test leakage; and we evaluate on 15 large, diverse public forecast competition datasets. Our empirical analysis reveals that statistical models’ accuracy is frequently underreported. Notably, we confirm that statistical models and their ensembles consistently outperform existing FFMs by more than 8.2% in sCRPS, and by more than 20% MASE, across datasets. However, we also find that synthetic dataset pre-training does improve the accuracy of a FFM by 7% percent. Read More
Differentiated Directional Intervention A Framework for Evading LLM Safety Alignmentcs.AI updates on arXiv.org arXiv:2511.06852v2 Announce Type: replace-cross
Abstract: Safety alignment instills in Large Language Models (LLMs) a critical capacity to refuse malicious requests. Prior works have modeled this refusal mechanism as a single linear direction in the activation space. We posit that this is an oversimplification that conflates two functionally distinct neural processes: the detection of harm and the execution of a refusal. In this work, we deconstruct this single representation into a Harm Detection Direction and a Refusal Execution Direction. Leveraging this fine-grained model, we introduce Differentiated Bi-Directional Intervention (DBDI), a new white-box framework that precisely neutralizes the safety alignment at critical layer. DBDI applies adaptive projection nullification to the refusal execution direction while suppressing the harm detection direction via direct steering. Extensive experiments demonstrate that DBDI outperforms prominent jailbreaking methods, achieving up to a 97.88% attack success rate on models such as Llama-2. By providing a more granular and mechanistic framework, our work offers a new direction for the in-depth understanding of LLM safety alignment.
arXiv:2511.06852v2 Announce Type: replace-cross
Abstract: Safety alignment instills in Large Language Models (LLMs) a critical capacity to refuse malicious requests. Prior works have modeled this refusal mechanism as a single linear direction in the activation space. We posit that this is an oversimplification that conflates two functionally distinct neural processes: the detection of harm and the execution of a refusal. In this work, we deconstruct this single representation into a Harm Detection Direction and a Refusal Execution Direction. Leveraging this fine-grained model, we introduce Differentiated Bi-Directional Intervention (DBDI), a new white-box framework that precisely neutralizes the safety alignment at critical layer. DBDI applies adaptive projection nullification to the refusal execution direction while suppressing the harm detection direction via direct steering. Extensive experiments demonstrate that DBDI outperforms prominent jailbreaking methods, achieving up to a 97.88% attack success rate on models such as Llama-2. By providing a more granular and mechanistic framework, our work offers a new direction for the in-depth understanding of LLM safety alignment. Read More
Think Before You Retrieve: Learning Test-Time Adaptive Search with Small Language Modelscs.AI updates on arXiv.org arXiv:2511.07581v1 Announce Type: new
Abstract: Effective information retrieval requires reasoning over partial evidence and refining strategies as information emerges. Yet current approaches fall short: neural retrievers lack reasoning capabilities, large language models (LLMs) provide semantic depth but at prohibitive cost, and query rewriting or decomposition limits improvement to static transformations. As a result, existing methods fail to capture the iterative dynamics of exploration, feedback, and revision that complex user queries demand. We introduce Orion, a training framework that enables compact models (350M-1.2B parameters) to perform iterative retrieval through learned search strategies. Orion combines: (1) synthetic trajectory generation and supervised fine-tuning to encourage diverse exploration patterns in models, (2) reinforcement learning (RL) that rewards effective query refinement and backtracking behaviors, and (3) inference-time beam search algorithms that exploit the self-reflection capabilities learned during RL. Despite using only 3% of the training data available, our 1.2B model achieves 77.6% success on SciFact (vs. 72.6% for prior retrievers), 25.2% on BRIGHT (vs. 22.1%), 63.2% on NFCorpus (vs. 57.8%), and remains competitive on FEVER, HotpotQA, and MSMarco. It outperforms retrievers up to 200-400x larger on five of six benchmarks. These findings suggest that retrieval performance can emerge from learned strategies, not just model scale, when models are trained to search, reflect, and revise.
arXiv:2511.07581v1 Announce Type: new
Abstract: Effective information retrieval requires reasoning over partial evidence and refining strategies as information emerges. Yet current approaches fall short: neural retrievers lack reasoning capabilities, large language models (LLMs) provide semantic depth but at prohibitive cost, and query rewriting or decomposition limits improvement to static transformations. As a result, existing methods fail to capture the iterative dynamics of exploration, feedback, and revision that complex user queries demand. We introduce Orion, a training framework that enables compact models (350M-1.2B parameters) to perform iterative retrieval through learned search strategies. Orion combines: (1) synthetic trajectory generation and supervised fine-tuning to encourage diverse exploration patterns in models, (2) reinforcement learning (RL) that rewards effective query refinement and backtracking behaviors, and (3) inference-time beam search algorithms that exploit the self-reflection capabilities learned during RL. Despite using only 3% of the training data available, our 1.2B model achieves 77.6% success on SciFact (vs. 72.6% for prior retrievers), 25.2% on BRIGHT (vs. 22.1%), 63.2% on NFCorpus (vs. 57.8%), and remains competitive on FEVER, HotpotQA, and MSMarco. It outperforms retrievers up to 200-400x larger on five of six benchmarks. These findings suggest that retrieval performance can emerge from learned strategies, not just model scale, when models are trained to search, reflect, and revise. Read More
Reference-Guided Verdict: LLMs-as-Judges in Automatic Evaluation of Free-Form QAcs.AI updates on arXiv.org arXiv:2408.09235v3 Announce Type: replace-cross
Abstract: The emergence of Large Language Models (LLMs) as chat assistants capable of generating human-like conversations has amplified the need for robust evaluation methods, particularly for open-ended tasks. Conventional metrics such as EM and F1, while useful, are inadequate for capturing the full semantics and contextual depth of such generative outputs. We propose a reference-guided verdict method that automates the evaluation process by leveraging multiple LLMs as judges. Through experiments on free-form question-answering tasks, we demonstrate that combining multiple models improves the reliability and accuracy of evaluations, especially in tasks where a single model may struggle. The results indicate a strong correlation with human evaluations, establishing the proposed method as a reliable alternative to traditional metrics.
arXiv:2408.09235v3 Announce Type: replace-cross
Abstract: The emergence of Large Language Models (LLMs) as chat assistants capable of generating human-like conversations has amplified the need for robust evaluation methods, particularly for open-ended tasks. Conventional metrics such as EM and F1, while useful, are inadequate for capturing the full semantics and contextual depth of such generative outputs. We propose a reference-guided verdict method that automates the evaluation process by leveraging multiple LLMs as judges. Through experiments on free-form question-answering tasks, we demonstrate that combining multiple models improves the reliability and accuracy of evaluations, especially in tasks where a single model may struggle. The results indicate a strong correlation with human evaluations, establishing the proposed method as a reliable alternative to traditional metrics. Read More
Training Language Models to Explain Their Own Computationscs.AI updates on arXiv.org arXiv:2511.08579v1 Announce Type: cross
Abstract: Can language models (LMs) learn to faithfully describe their internal computations? Are they better able to describe themselves than other models? We study the extent to which LMs’ privileged access to their own internals can be leveraged to produce new techniques for explaining their behavior. Using existing interpretability techniques as a source of ground truth, we fine-tune LMs to generate natural language descriptions of (1) the information encoded by LM features, (2) the causal structure of LMs’ internal activations, and (3) the influence of specific input tokens on LM outputs. When trained with only tens of thousands of example explanations, explainer models exhibit non-trivial generalization to new queries. This generalization appears partly attributable to explainer models’ privileged access to their own internals: using a model to explain its own computations generally works better than using a *different* model to explain its computations (even if the other model is significantly more capable). Our results suggest not only that LMs can learn to reliably explain their internal computations, but that such explanations offer a scalable complement to existing interpretability methods.
arXiv:2511.08579v1 Announce Type: cross
Abstract: Can language models (LMs) learn to faithfully describe their internal computations? Are they better able to describe themselves than other models? We study the extent to which LMs’ privileged access to their own internals can be leveraged to produce new techniques for explaining their behavior. Using existing interpretability techniques as a source of ground truth, we fine-tune LMs to generate natural language descriptions of (1) the information encoded by LM features, (2) the causal structure of LMs’ internal activations, and (3) the influence of specific input tokens on LM outputs. When trained with only tens of thousands of example explanations, explainer models exhibit non-trivial generalization to new queries. This generalization appears partly attributable to explainer models’ privileged access to their own internals: using a model to explain its own computations generally works better than using a *different* model to explain its computations (even if the other model is significantly more capable). Our results suggest not only that LMs can learn to reliably explain their internal computations, but that such explanations offer a scalable complement to existing interpretability methods. Read More
EdgeRunner 20B: Military Task Parity with GPT-5 while Running on the Edgecs.AI updates on arXiv.org arXiv:2510.26550v2 Announce Type: replace
Abstract: We present EdgeRunner 20B, a fine-tuned version of gpt-oss-20b optimized for military tasks. EdgeRunner 20B was trained on 1.6M high-quality records curated from military documentation and websites. We also present four new tests sets: (a) combat arms, (b) combat medic, (c) cyber operations, and (d) mil-bench-5k (general military knowledge). On these military test sets, EdgeRunner 20B matches or exceeds GPT-5 task performance with 95%+ statistical significance, except for the high reasoning setting on the combat medic test set and the low reasoning setting on the mil-bench-5k test set. Versus gpt-oss-20b, there is no statistically-significant regression on general-purpose benchmarks like ARC-C, GPQA Diamond, GSM8k, IFEval, MMLU Pro, or TruthfulQA, except for GSM8k in the low reasoning setting. We also present analyses on hyperparameter settings, cost, and throughput. These findings show that small, locally-hosted models are ideal solutions for data-sensitive operations such as in the military domain, allowing for deployment in air-gapped edge devices.
arXiv:2510.26550v2 Announce Type: replace
Abstract: We present EdgeRunner 20B, a fine-tuned version of gpt-oss-20b optimized for military tasks. EdgeRunner 20B was trained on 1.6M high-quality records curated from military documentation and websites. We also present four new tests sets: (a) combat arms, (b) combat medic, (c) cyber operations, and (d) mil-bench-5k (general military knowledge). On these military test sets, EdgeRunner 20B matches or exceeds GPT-5 task performance with 95%+ statistical significance, except for the high reasoning setting on the combat medic test set and the low reasoning setting on the mil-bench-5k test set. Versus gpt-oss-20b, there is no statistically-significant regression on general-purpose benchmarks like ARC-C, GPQA Diamond, GSM8k, IFEval, MMLU Pro, or TruthfulQA, except for GSM8k in the low reasoning setting. We also present analyses on hyperparameter settings, cost, and throughput. These findings show that small, locally-hosted models are ideal solutions for data-sensitive operations such as in the military domain, allowing for deployment in air-gapped edge devices. Read More
Remodeling Semantic Relationships in Vision-Language Fine-Tuningcs.AI updates on arXiv.org arXiv:2511.08238v1 Announce Type: cross
Abstract: Vision-language fine-tuning has emerged as an efficient paradigm for constructing multimodal foundation models. While textual context often highlights semantic relationships within an image, existing fine-tuning methods typically overlook this information when aligning vision and language, thus leading to suboptimal performance. Toward solving this problem, we propose a method that can improve multimodal alignment and fusion based on both semantics and relationships.Specifically, we first extract multilevel semantic features from different vision encoder to capture more visual cues of the relationships. Then, we learn to project the vision features to group related semantics, among which are more likely to have relationships. Finally, we fuse the visual features with the textual by using inheritable cross-attention, where we globally remove the redundant visual relationships by discarding visual-language feature pairs with low correlation. We evaluate our proposed method on eight foundation models and two downstream tasks, visual question answering and image captioning, and show that it outperforms all existing methods.
arXiv:2511.08238v1 Announce Type: cross
Abstract: Vision-language fine-tuning has emerged as an efficient paradigm for constructing multimodal foundation models. While textual context often highlights semantic relationships within an image, existing fine-tuning methods typically overlook this information when aligning vision and language, thus leading to suboptimal performance. Toward solving this problem, we propose a method that can improve multimodal alignment and fusion based on both semantics and relationships.Specifically, we first extract multilevel semantic features from different vision encoder to capture more visual cues of the relationships. Then, we learn to project the vision features to group related semantics, among which are more likely to have relationships. Finally, we fuse the visual features with the textual by using inheritable cross-attention, where we globally remove the redundant visual relationships by discarding visual-language feature pairs with low correlation. We evaluate our proposed method on eight foundation models and two downstream tasks, visual question answering and image captioning, and show that it outperforms all existing methods. Read More
Building a Gmail Inbox Management Agent in n8nKDnuggets Learn to create an intelligent email automation system that analyzes, scores, and routes messages based on priority.
Learn to create an intelligent email automation system that analyzes, scores, and routes messages based on priority. Read More
