최근 비전-언어 모델(VLM)은 이미지로부터 실행 가능한 CAD 프로그램을 생성할 수 있게 되었지만, 기존 방식들은 주로 대략적이고 범용적인 3D 객체를 대상으로 하며 산업 기계 설계에서 요구되는 세밀한 기하학적 형상과 밀리미터 단위의 공차를 거의 다루지 못합니다. 본 논문에서는 산업 제조 데이터로부터 실행 가능한 CAD 생성을 수행하는 VLM을 평가하기 위한 최초의 백만 단위 규모 벤치마크인 OmniMech를 소개합니다. OmniMech는 251,000개 이상의 치수 및 공차가 완벽하게 기입된 2D 직교 투상도면(orthographic drawing)과 함께, 이에 대응하는 네이티브 CAD 모델, 다중 시점 렌더링, 메시, STEP 및 B-rep 표현, 풍부한 의미론적 주석을 포함합니다. 이 벤치마크는 네 가지 태스크로 구성됩니다: (1) 공학 도면으로부터 파라메트릭 CAD 프로그램 합성; (2) 기하학적 및 구조적으로 일관된 재구성을 위한 도면-3D 추론; (3) 치수, 기호, 특징(feature) 지시선 및 제조 제약 조건에 기반한 주석 그라운딩 추론; (4) 시각화, 측정, CAD 실행 및 검증 도구를 활용하는 도구 증강 에이전트 추론. 실험 결과에 따르면 현재의 VLM 및 CAD 특화 모델들은 여전히 실행 가능한 프로그램 합성, 세밀한 3D 재구성, 신뢰할 수 있는 치수 및 공차 준수에 어려움을 겪고 있습니다. 향후 연구를 지원하기 위해 벤치마크 데이터, 평가 코드 및 도구 인터페이스를 공개할 예정입니다.
Various families of malware use domain generation algorithms (DGAs) to generate a large number of pseudo-random domain names to connect to a command and control (C&C) server. In order to block DGA C&C traffic, security organizations must first discover the algorithm by reverse engineering malware samples, then generating a list of domains for a given seed. The domains are then either preregistered or published in a DNS blacklist. This process is not only tedious, but can be readily circumvented by malware authors using a large number of seeds in algorithms with multivariate recurrence properties (e.g., banjori) or by using a dynamic list of seeds (e.g., bedep). Another technique to stop malware from using DGAs is to intercept DNS queries on a network and predict whether domains are DGA generated. Such a technique will alert network administrators to the presence of malware on their networks. In addition, if the predictor can also accurately predict the family of DGAs, then network administrators can also be alerted to the type of malware that is on their networks. This paper presents a DGA classifier that leverages long short-term memory (LSTM) networks to predict DGAs and their respective families without the need for a priori feature extraction. Results are significantly better than state-of-the-art techniques, providing 0.9993 area under the receiver operating characteristic curve for binary classification and a micro-averaged F1 score of 0.9906. In other terms, the LSTM technique can provide a 90% detection rate with a 1:10000 false positive (FP) rate---a twenty times FP improvement over comparable methods. Experiments in this paper are run on open datasets and code snippets are provided to reproduce the results.
Large language models (LLMs) have achieved record adoption in a short period of time across many different sectors including high importance areas such as education [4] and healthcare [23]. LLMs are open-ended models trained on diverse data without being tailored for specific downstream tasks, enabling broad applicability across various domains. They are commonly used for text generation, but also widely used to assist with code generation [3], and even analysis of security information, as Microsoft Security Copilot demonstrates [18]. Traditional Machine Learning (ML) models are vulnerable to adversarial attacks [9]. So the concerns on the potential security implications of such wide scale adoption of LLMs have led to the creation of this working group on the security of LLMs. During the Dagstuhl seminar on "Network Attack Detection and Defense - AI-Powered Threats and Responses", the working group discussions focused on the vulnerability of LLMs to adversarial attacks, rather than their potential use in generating malware or enabling cyberattacks. Although we note the potential threat represented by the latter, the role of the LLMs in such uses is mostly as an accelerator for development, similar to what it is in benign use. To make the analysis more specific, the working group employed ChatGPT as a concrete example of an LLM and addressed the following points, which also form the structure of this report: 1. How do LLMs differ in vulnerabilities from traditional ML models? 2. What are the attack objectives in LLMs? 3. How complex it is to assess the risks posed by the vulnerabilities of LLMs? 4. What is the supply chain in LLMs, how data flow in and out of systems and what are the security implications? We conclude with an overview of open challenges and outlook.
Text-to-image models have achieved remarkable success in generating high-quality images from textual descriptions. However, they often struggle with "semantic drift," where the generated output fails to precisely align with complex or nuanced text prompts. While recent approaches have attempted to address semantic errors regarding attribute binding or object presence, there remains a gap for a more holistic method that addresses these issues by directly refining the text embeddings of the initial user prompt. In this work, we introduce Orthogonal Drift Correction (ODC), an inference-time guidance method designed to mitigate semantic drift without requiring model retraining or additional user inputs. ODC guides image generation through a two-stage process. In the first stage, it identifies the semantic drift by evaluating the initially generated image against the user prompt in a shared vision-language embedding space. It then isolates the component of this drift vector that is orthogonal to the prompt's direction and translates it back into text via a vocabulary-based surrogate mechanism. In the second stage, it produces refined text conditioning for a second generation pass by feeding both the initial text embedding and the re-embedded drift representation into an adaptive rank-reduced concept removal module. Our experiments demonstrate that ODC consistently improves prompt-image alignment, increasing CLIP scores by up to 5.4% over the baseline on the COALBench-300 and PartiPrompts benchmarks. Furthermore, on the T2I-CompBench complex subset, ODC improves the average 3-in-1 score by 4.5% over the competing methods. As a plug-and-play module, ODC offers a practical and efficient method for improving the reliability of text-to-image models. Code is available at https://github.com/kasrakamali/ODC.
프라이버시 향상, 개인화, 지연 시간 감소 등 다양한 이점으로 엣지의 모바일 플랫폼에 대규모 언어 모델(LLM)을 직접 배포하는 방식이 주목받고 있다. 그러나 LLM은 높은 계산 요구량을 갖고 있어, 자원이 제한된 모바일 및 엣지 플랫폼이 이를 충족하기 어렵다. 계산 자원 제한 외에도 모바일 및 엣지 시스템은 종종 소형 폼팩터를 갖고, 높은 프로세서 사용률로 발생하는 열을 배출할 물리적 수단(예: 팬)이 없어 스로틀링과 처리 성능 저하를 막기 어렵고, LLM은 이를 쉽게 유발할 수 있다. 이를 완화하기 위해 선행 연구들은 모바일 플랫폼에서 무거운 계산 작업의 전력 및 발열을 줄이기 위해 동적 전압 및 주파수 스케일링(DVFS)과 같은 다양한 전력 관리 전략을 제안했다. 최근 모바일 LLM에 맞춘 DVFS 방법도 제안되었으나, 이러한 방법은 대부분 하드웨어 매개변수와 프로세서 주파수 최적화에 초점을 두며, 일부 열 제약 시나리오에서는 한계를 보인다. 최근 머신러닝 연구의 진전에서, 고품질 생성을 유지하기 위해 모든 토큰이 전 깊이(full-depth) 추론을 필요로 하지 않는다는 핵심 통찰을 우리는 식별하고 이를 활용한다. 이에 영감을 받아, 우리는 PELM을 제present한다. PELM은 전통적인 DVFS 프로세서 주파수 조정에 두 가지 워크로드 특화 조절 요소, 즉 1) 추측 디코딩(speculative decoding)과 2) 가변 검증 깊이(variable verification depth)를 더해 최적화 공간을 다차원으로 확장하고, 보다 전력 효율적인 온디바이스 LLM 추론을 달성하는 솔루션이다. 다양한 하드웨어 플랫폼과 데이터셋에 걸친 광범위한 평가에서 PELM은 최신 전력 관리 방법 대비 우수한 성능을 보이며, 최대 23.1% 속도 향상과 52.4% 에너지 소비 감소를 달성하면서 유사한 수준의 태스크 성능을 유지한다. 소스 코드는 https://github.com/imec-nu/PELM 에서 이용할 수 있다.
Image-conditioned video models are used to show real places, so whether their output stays faithful to the place they were shown is a practical question. The text benchmarks we found that check fidelity against a reference do so at an anchor - an annotated region, typically the sign the task is about. We show that this choice hides most of the failure. We conditioned four models from three vendors on 48 street-level references of Tokyo storefronts and measured text fidelity at the annotated sign and in the periphery, the other text the reference covers. At the anchor, all four models are close to a no-model control (anchor text fidelity 0.90-0.94 against 0.996) and are not reliably separated from one another. In the periphery, 36-54% of generated text regions have no counterpart in the reference, against 1.6% for the control, although every prompt forbade new readable text. Two blinded raters agree on which regions are fabricated (kappa = 0.83-0.96), the result held in preregistered rounds on two vendors whose videos the measure had never seen (42% and 56%), and VBench-I2V's reference-based consistency scores show no evidence of tracking it in our sample. Most fabricated text is legible-looking text where the reference shows none or cannot be read - a form that string-matching metrics discard, which is why our own first measure missed it. A preregistered Latin-script replication in London missed its bar in a first round and met it in an independent second round. We release code, annotations, per-video scores, ratings and every preregistration. Code and data: https://github.com/geoywang/refanchor-bench
시뮬레이션 기반 최적화(SBO)는 확률적 동역학 하에서 실행 가능한 정책을 평가하지만, 대부분의 방법은 시뮬레이터를 블랙박스로 취급합니다. 즉, 집계 점수로 후보의 순위를 매길 뿐 후보가 실패하는 이유나 어떤 정책 로직을 변경해야 하는지는 밝히지 못합니다. 본 논문에서는 선택을 위한 반복 시뮬레이션과 진단을 위한 이벤트 수준 추적(trace)을 활용하는 대규모 언어 모델(LLM) 유도 휴리스틱 설계 프레임워크를 제시합니다. 각 현행 정책은 여러 번의 반복 수행을 통해 평가되며, 가장 점수가 낮은 실행을 재생하여 쿼리 가능한 추적을 생성합니다. 관리자 에이전트(manager agent)는 이 증거로부터 병목 가설을 수립하고, 편집 에이전트(editing agent)들은 병렬로 코드 수준의 수정을 구현합니다. 실행 점검과 반복 평가를 거쳐, 역대 최고 선택 방식(best-so-far selection)을 통해 개선된 것만 유지합니다. LLM 수정은 평가 배치 사이에서 발생하며, 각 시뮬레이션 실행은 고정된 정책으로 제어됩니다. 저자들은 동적 생산 및 무인 운반차(AGV) 스케줄링의 이산 사건 시뮬레이션에서 본 프레임워크를 평가합니다. Gemini-3.1-Pro를 사용한 5회의 독립적인 최적화 실행 전반에 걸쳐, 시뮬레이터의 0-100 척도에서 최종 평균 점수는 평균 77.51점을 기록했습니다. 가장 점수가 높은 실행에서 추적 기반 진단은 선제적 충전, 거리 인지 AGV 할당, 디스패치 우선순위 재조정을 유도하여 역대 최고 평균 점수를 62.49점에서 78.61점으로 끌어올렸습니다. 일치된 100개의 시드에서, 최고의 최종 정책은 모든 시드에 걸쳐 대표적인 롤링-MILP, 규칙 기반, 메타휴리스틱 정책보다 높은 점수를 기록했으며, 재최적화 없이 무작위 고장 상황에서도 우위를 유지했습니다. 더 긴 기간(horizon) 및 가변 주문 도착 간격에 대한 별도의 재최적화 후에도, 도출된 정책들은 다시 모든 기준선을 능가했습니다. 두 가지 LLM 백본을 사용한 소거 연구(ablation) 결과, 병렬 후보 생성이나 추적 데이터베이스 접근 중 하나만 제거해도 최종 평균 점수가 감소함을 확인했습니다. 이러한 결과는 시뮬레이션 추적이 복잡한 시뮬레이션 기반 스케줄링에서 목표 지향적인 코드 수준의 정책 개선을 이끌 수 있음을 보여줍니다.
In this paper, we present a customizable datacentric system that automatically generates common misspellings for complex health-related terms. The spelling variant generator relies on a dense vector model learned from large unlabeled text, which is used to find semantically close terms to the original/seed keyword, followed by the filtering of terms that are lexically dissimilar beyond a given threshold. The process is executed recursively, converging when no new terms similar (lexically and semantically) to the seed keyword are found. Weighting of intra-word character sequence similarities allows further problem-specific customization of the system. On a dataset prepared for this study, our system outperforms the current state-of-the-art for medication name variant generation with best F1-score of 0.69 and F1/4-score of 0.78. Extrinsic evaluation of the system on a set of cancer-related terms showed an increase of over 67% in retrieval rate from Twitter posts when the generated variants are included. Our proposed spelling variant generator has several advantages over the current state-of-the-art and other types of variant generators-(i) it is capable of filtering out lexically similar but semantically dissimilar terms, (ii) the number of variants generated is low as many low-frequency and ambiguous misspellings are filtered out, and (iii) the system is fully automatic, customizable and easily executable. While the base system is fully unsupervised, we show how supervision maybe employed to adjust weights for task-specific customization. The performance and significant relative simplicity of our proposed approach makes it a much needed misspelling generation resource for health-related text mining from noisy sources. The source code for the system has been made publicly available for research purposes.
Genetic Programming (GP) is an heuristic method that can be applied to many Machine Learning, Optimization and Engineering problems. In particular, it has been widely used in Software Engineering for Test-case generation, Program Synthesis and Improvement of Software (GI). Grammar-Guided Genetic Programming (GGGP) approaches allow the user to refine the domain of valid program solutions. Backus Normal Form is the most popular interface for describing Context-Free Grammars (CFG) for GGGP. BNF and its derivatives have the disadvantage of interleaving the grammar language and the target language of the program. We propose to embed the grammar as an internal Domain-Specific Language in the host language of the framework. This approach has the same expressive power as BNF and EBNF while using the host language type-system to take advantage of all the existing tooling: linters, formatters, type-checkers, autocomplete, and legacy code support. These tools have a practical utility in designing software in general, and GP systems in particular. We also present Meta-Handlers, user-defined overrides of the tree-generation system. This technique extends our object-oriented encoding with more practicability and expressive power than existing CFG approaches, achieving the same expressive power of Attribute Grammars, but without the grammar vs target language duality. Furthermore, we evidence that this approach is feasible, showing an example Python implementation as proof. We also compare our approach against textual BNF-representations w.r.t. expressive power and ergonomics. These advantages do not come at the cost of performance, as shown by our empirical evaluation on 5 benchmarks of our example implementation against PonyGE2. We conclude that our approach has better ergonomics with the same expressive power and performance of textual BNF-based grammar encodings.
Big Data empowers the farming community with the information needed to optimize resource usage, increase productivity, and enhance the sustainability of agricultural practices. The use of Big Data in farming requires the collection and analysis of data from various sources such as sensors, satellites, and farmer surveys. While Big Data can provide the farming community with valuable insights and improve efficiency, there is significant concern regarding the security of this data as well as the privacy of the participants. Privacy regulations, such as the EU GDPR, the EU Code of Conduct on agricultural data sharing by contractual agreement, and the proposed EU AI law, have been created to address the issue of data privacy and provide specific guidelines on when and how data can be shared between organizations. To make confidential agricultural data widely available for Big Data analysis without violating the privacy of the data subjects, we consider privacy-preserving methods of data sharing in agriculture. Deep learning-based synthetic data generation has been proposed for privacy-preserving data sharing. However, there is a lack of compliance with documented data privacy policies in such privacy-preserving efforts. In this study, we propose a novel framework for enforcing privacy policy rules in privacy-preserving data generation algorithms. We explore several available agricultural codes of conduct, extract knowledge related to the privacy constraints in data, and use the extracted knowledge to define privacy bounds in a privacy-preserving generative model. We use our framework to generate synthetic agricultural data and present experimental results that demonstrate the utility of the synthetic dataset in downstream tasks. We also show that our framework can evade potential threats and secure data based on applicable regulatory policy rules.
Despite the transformative impact of Large Language Models (LLMs) across critical domains such as healthcare, customer service, and business marketing, their integration into Open Radio Access Networks (O-RAN) remains limited. This gap is primarily due to the absence of domain-specific foundational models, with existing solutions often relying on general-purpose LLMs that fail to address the unique challenges and technical intricacies of O-RAN. To bridge this gap, we introduce ORANSight-2.0 (O-RAN Insights), a pioneering initiative aimed at developing specialized foundational LLMs tailored for O-RAN. Built on 18 LLMs spanning five open-source LLM frameworks, ORANSight-2.0 fine-tunes models ranging from 1 to 70B parameters, significantly reducing reliance on proprietary, closed-source models while enhancing performance for O-RAN. At the core of ORANSight-2.0 is RANSTRUCT, a novel Retrieval-Augmented Generation (RAG) based instruction-tuning framework that employs two LLM agents to create high-quality instruction-tuning datasets. The generated dataset is then used to fine-tune the 18 pre-trained open-source LLMs via QLoRA. To evaluate ORANSight-2.0, we introduce srsRANBench, a novel benchmark designed for code generation and codebase understanding in the context of srsRAN, a widely used 5G O-RAN stack. We also leverage ORANBench13K, an existing benchmark for assessing O-RAN-specific knowledge. Our comprehensive evaluations demonstrate that ORANSight-2.0 models outperform general-purpose and closed-source models, such as ChatGPT-4o and Gemini, by 5.421% on ORANBench and 18.465% on srsRANBench, achieving superior performance while maintaining lower computational and energy costs. We also experiment with RAG-augmented variants of ORANSight-2.0 LLMs and thoroughly evaluate their energy characteristics, demonstrating costs for training, standard inference, and RAG-augmented inference.
Tool-augmented LLM agents tend to call tools indiscriminately, even when the model can answer directly. Each unnecessary call wastes API fees and latency, yet no existing benchmark systematically studies when a tool call is actually needed. We propose When2Tool, a benchmark of 18 environments (15 single-hop, 3 multi-hop) spanning three categories of tool necessity -- computational scale, knowledge boundaries, and execution reliability -- each with controlled difficulty levels that create a clear decision boundary between tool-necessary and tool-unnecessary tasks. We evaluate two families of training-free baselines: Prompt-only (varying the prompt to discourage unnecessary calls) and Reason-then-Act (requiring the model to reason about tool necessity before acting). Both provide limited control: Prompt-only suppresses necessary calls alongside unnecessary ones, and Reason-then-Act still incurs a disproportionate accuracy cost on hard tasks. To understand why these baselines fail, we probe the models' hidden states and find that tool necessity is linearly decodable from the pre-generation representation with AUROC 0.89--0.96 across six models, substantially exceeding the model's own verbalized reasoning. This reveals that models already know when tools are needed, but fail to act on this knowledge during generation. Building on this finding, we propose Probe&Prefill, which uses a lightweight linear probe to read the hidden-state signal and prefills the model's response with a steering sentence. Across all models tested, Probe&Prefill reduces tool calls by 38\% with only 1.7\% accuracy loss, while the best baseline at comparable accuracy only reduces 6% of tool calls, or achieves a similar tool call reduction but incurs a 5$\times$ higher accuracy loss. Our code is available at https://github.com/Trustworthy-ML-Lab/when2tool
Current end-to-end grasp planning methods propose grasps in the order of seconds that attain high grasp success rates on a diverse set of objects, but often by constraining the workspace to top-grasps. In this work, we present a method that allows end-to-end top-grasp planning methods to generate full six-degree-of-freedom grasps using a single RGB-D view as input. This is achieved by estimating the complete shape of the object to be grasped, then simulating different viewpoints of the object, passing the simulated viewpoints to an end-to-end grasp generation method, and finally executing the overall best grasp. The method was experimentally validated on a Franka Emika Panda by comparing 429 grasps generated by the state-of-the-art Fully Convolutional Grasp Quality CNN, both on simulated and real camera images. The results show statistically significant improvements in terms of grasp success rate when using simulated images over real camera images, especially when the real camera viewpoint is angled. Code and video are available at https://irobotics.aalto.fi/beyond-top-grasps-through-scene-completion/.
We propose a novel theoretical framework to understand contrastive self-supervised learning (SSL) methods that employ dual pairs of deep ReLU networks (e.g., SimCLR). First, we prove that in each SGD update of SimCLR with various loss functions, including simple contrastive loss, soft Triplet loss and InfoNCE loss, the weights at each layer are updated by a \emph{covariance operator} that specifically amplifies initial random selectivities that vary across data samples but survive averages over data augmentations. To further study what role the covariance operator plays and which features are learned in such a process, we model data generation and augmentation processes through a \emph{hierarchical latent tree model} (HLTM) and prove that the hidden neurons of deep ReLU networks can learn the latent variables in HLTM, despite the fact that the network receives \emph{no direct supervision} from these unobserved latent variables. This leads to a provable emergence of hierarchical features through the amplification of initially random selectivities through contrastive SSL. Extensive numerical studies justify our theoretical findings. Code is released in https://github.com/facebookresearch/luckmatters/tree/master/ssl.
Recently, GAN inversion methods combined with Contrastive Language-Image Pretraining (CLIP) enables zero-shot image manipulation guided by text prompts. However, their applications to diverse real images are still difficult due to the limited GAN inversion capability. Specifically, these approaches often have difficulties in reconstructing images with novel poses, views, and highly variable contents compared to the training data, altering object identity, or producing unwanted image artifacts. To mitigate these problems and enable faithful manipulation of real images, we propose a novel method, dubbed DiffusionCLIP, that performs text-driven image manipulation using diffusion models. Based on full inversion capability and high-quality image generation power of recent diffusion models, our method performs zero-shot image manipulation successfully even between unseen domains and takes another step towards general application by manipulating images from a widely varying ImageNet dataset. Furthermore, we propose a novel noise combination method that allows straightforward multi-attribute manipulation. Extensive experiments and human evaluation confirmed robust and superior manipulation performance of our methods compared to the existing baselines. Code is available at https://github.com/gwang-kim/DiffusionCLIP.git.
Sequential recommendation systems are integral to discerning temporal user preferences. Yet, the task of learning from abbreviated user interaction sequences poses a notable challenge. Data augmentation has been identified as a potent strategy to enhance the informational richness of these sequences. Traditional augmentation techniques, such as item randomization, may disrupt the inherent temporal dynamics. Although recent advancements in reverse chronological pseudo-item generation have shown promise, they can introduce temporal discrepancies when assessed in a natural chronological context. In response, we introduce a sophisticated approach, Bidirectional temporal data Augmentation with pre-training (BARec). Our approach leverages bidirectional temporal augmentation and knowledge-enhanced fine-tuning to synthesize authentic pseudo-prior items that retain user preferences and capture deeper item semantic correlations, thus boosting the model's expressive power. Our comprehensive experimental analysis on five benchmark datasets confirms the superiority of BARec across both short and elongated sequence contexts. Moreover, theoretical examination and case study offer further insight into the model's logical processes and interpretability. The source code for our study is publicly available at https://github.com/juyongjiang/BARec.
We introduce AiD Regen, a novel system that generates 3D wound models combining 2D semantic segmentation with 3D reconstruction so that they can be printed via 3D bio-printers during the surgery to treat diabetic foot ulcers (DFUs). AiD Regen seamlessly binds the full pipeline, which includes RGB-D image capturing, semantic segmentation, boundary-guided point-cloud processing, 3D model reconstruction, and 3D printable G-code generation, into a single system that can be used out of the box. We developed a multi-stage data preprocessing method to handle small and unbalanced DFU image datasets. AiD Regen's human-in-the-loop machine learning interface enables clinicians to not only create 3D regenerative patches with just a few touch interactions but also customize and confirm wound boundaries. As evidenced by our experiments, our model outperforms prior wound segmentation models and our reconstruction algorithm is capable of generating 3D wound models with compelling accuracy. We further conducted a case study on a real DFU patient and demonstrated the effectiveness of AiD Regen in treating DFU wounds.
This paper presents an unpaired method for creating line drawings from photographs. Current methods often rely on high quality paired datasets to generate line drawings. However, these datasets often have limitations due to the subjects of the drawings belonging to a specific domain, or in the amount of data collected. Although recent work in unsupervised image-to-image translation has shown much progress, the latest methods still struggle to generate compelling line drawings. We observe that line drawings are encodings of scene information and seek to convey 3D shape and semantic meaning. We build these observations into a set of objectives and train an image translation to map photographs into line drawings. We introduce a geometry loss which predicts depth information from the image features of a line drawing, and a semantic loss which matches the CLIP features of a line drawing with its corresponding photograph. Our approach outperforms state-of-the-art unpaired image translation and line drawing generation methods on creating line drawings from arbitrary photographs. For code and demo visit our webpage carolineec.github.io/informative_drawings
Recently, large language models (LLMs) (e.g., GPT-4) have demonstrated impressive general-purpose task-solving abilities, including the potential to approach recommendation tasks. Along this line of research, this work aims to investigate the capacity of LLMs that act as the ranking model for recommender systems. We first formalize the recommendation problem as a conditional ranking task, considering sequential interaction histories as conditions and the items retrieved by other candidate generation models as candidates. To solve the ranking task by LLMs, we carefully design the prompting template and conduct extensive experiments on two widely-used datasets. We show that LLMs have promising zero-shot ranking abilities but (1) struggle to perceive the order of historical interactions, and (2) can be biased by popularity or item positions in the prompts. We demonstrate that these issues can be alleviated using specially designed prompting and bootstrapping strategies. Equipped with these insights, zero-shot LLMs can even challenge conventional recommendation models when ranking candidates are retrieved by multiple candidate generators. The code and processed datasets are available at https://github.com/RUCAIBox/LLMRank.
Diffusion models have recently demonstrated an impressive ability to address inverse problems in an unsupervised manner. While existing methods primarily focus on modifying the posterior sampling process, the potential of the forward process remains largely unexplored. In this work, we propose Shortcut Sampling for Diffusion(SSD), a novel approach for solving inverse problems in a zero-shot manner. Instead of initiating from random noise, the core concept of SSD is to find a specific transitional state that bridges the measurement image y and the restored image x. By utilizing the shortcut path of "input - transitional state - output", SSD can achieve precise restoration with fewer steps. To derive the transitional state during the forward process, we introduce Distortion Adaptive Inversion. Moreover, we apply back projection as additional consistency constraints during the generation process. Experimentally, we demonstrate SSD's effectiveness on multiple representative IR tasks. Our method achieves competitive results with only 30 NFEs compared to state-of-the-art zero-shot methods(100 NFEs) and outperforms them with 100 NFEs in certain tasks. Code is available at https://github.com/GongyeLiu/SSD