Author name cluster

Lingfeng Shen

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

8 papers

2 author rows

NeurIPS Conference 2025 Conference Paper

MIR-Bench: Can Your LLM Recognize Complicated Patterns via Many-Shot In-Context Reasoning?

Kai Yan
Zhan Ling
Kang Liu
Yifan Yang
Ting-Han Fan
Lingfeng Shen
Zhengyin Du
Jiecao Chen

The ability to recognize patterns from examples and apply them to new ones is a primal ability for general intelligence, and is widely studied by psychology and AI researchers. Many benchmarks have been proposed to measure such ability for Large Language Models (LLMs); however, they focus on few-shot (usually <10) setting and lack evaluation for aggregating many pieces of information from long contexts. On the other hand, the ever-growing context length of LLMs have brought forth the novel paradigm of many-shot In-Context Learning (ICL), which addresses new tasks with hundreds to thousands of examples without expensive and inefficient fine-tuning. However, many-shot evaluations often focus on classification, and popular long-context LLM tasks such as Needle-In-A-Haystack (NIAH) seldom require complicated intelligence for integrating many pieces of information. To fix the issues from both worlds, we propose MIR-Bench, the first many-shot in-context reasoning benchmark for pattern recognition that asks LLM to predict output via input-output examples from underlying functions with diverse data format. Based on MIR-Bench, we study many novel problems for many-shot in-context reasoning, and acquired many insightful findings including scaling effect, robustness, inductive vs. transductive reasoning, retrieval Augmented Generation (RAG), coding for inductive reasoning, cross-domain generalizability, etc. Our dataset is available at https: //huggingface. co/datasets/kaiyan289/MIR-Bench.

PDF Details

TMLR Journal 2025 Journal Article

Reliable and Responsible Foundation Models

Xinyu Yang
Junlin Han
Rishi Bommasani
Jinqi Luo
Wenjie Qu
Wangchunshu Zhou
Adel Bibi
Xiyao Wang

Foundation models, including Large Language Models (LLMs), Multimodal Large Language Models (MLLMs), Image Generative Models (i.e, Text-to-Image Models and Image-Editing Models), and Video Generative Models, have become essential tools with broad applications across various domains such as law, medicine, education, finance, and beyond. As these models see increasing real-world deployment, ensuring their reliability and responsibility has become critical for academia, industry, and government. This survey addresses the reliable and responsible development of foundation models. We explore critical issues, including bias and fairness, security and privacy, uncertainty, explainability, and distribution shift. Our research also covers model limitations, such as hallucinations, as well as methods like alignment and Artificial Intelligence-Generated Content (AIGC) detection. For each area, we review the current state of the field and outline concrete future research directions. Additionally, we discuss the intersections between these areas, highlighting their connections and shared challenges. We hope our survey fosters the development of foundation models that are not only powerful but also ethical, trustworthy, reliable, and socially responsible.

PDF Details

ICML Conference 2024 Conference Paper

Contrastive Preference Optimization: Pushing the Boundaries of LLM Performance in Machine Translation

Haoran Xu
Amr Sharaf
Yunmo Chen
Weiting Tan
Lingfeng Shen
Benjamin Van Durme
Kenton Murray
Young Jin Kim 0001

Moderate-sized large language models (LLMs) – those with 7B or 13B parameters – exhibit promising machine translation (MT) performance. However, they do not match the performance of state-of-the-art conventional encoder-decoder translation models or larger-scale LLMs such as GPT-4. In this study, we bridge this performance gap. We first assess the shortcomings of supervised fine-tuning for LLMs in the MT task, emphasizing the quality issues present in the reference data, despite being human-generated. Then, in contrast to supervised fine-tuning which mimics reference translations, we introduce Contrastive Preference Optimization (CPO), a novel approach that trains models to avoid generating adequate but not perfect translations. Applying CPO to ALMA models with only 22K parallel sentences and 0. 1% parameters yields significant improvements. The resulting model, called ALMA-R, can match or exceed the performance of the WMT competition winners and GPT-4 on WMT’21, WMT’22 and WMT’23 test datasets.

Details

NeurIPS Conference 2024 Conference Paper

DiffNorm: Self-Supervised Normalization for Non-autoregressive Speech-to-speech Translation

Weiting Tan
Jingyu Zhang
Lingfeng Shen
Daniel Khashabi
Philipp Koehn

Non-autoregressive Transformers (NATs) are recently applied in direct speech-to-speech translation systems, which convert speech across different languages without intermediate text data. Although NATs generate high-quality outputs and offer faster inference than autoregressive models, they tend to produce incoherent and repetitive results due to complex data distribution (e. g. , acoustic and linguistic variations in speech). In this work, we introduce DiffNorm, a diffusion-based normalization strategy that simplifies data distributions for training NAT models. After training with a self-supervised noise estimation objective, DiffNorm constructs normalized target data by denoising synthetically corrupted speech features. Additionally, we propose to regularize NATs with classifier-free guidance, improving model robustness and translation quality by randomly dropping out source information during training. Our strategies result in a notable improvement of about $+7$ ASR-BLEU for English-Spanish (En-Es) translation and $+2$ ASR-BLEU for English-French (En-Fr) on the CVSS benchmark, while attaining over $14\times$ speedup for En-Es and $5 \times$ speedup for En-Fr translations compared to autoregressive baselines.

PDF Details DOI

ICML Conference 2024 Conference Paper

Position: Do pretrained Transformers Learn In-Context by Gradient Descent?

Lingfeng Shen
Aayush Mishra
Daniel Khashabi

The emergence of In-Context Learning (ICL) in LLMs remains a remarkable phenomenon that is partially understood. To explain ICL, recent studies have created theoretical connections to Gradient Descent (GD). We ask, do such connections hold up in actual pre-trained language models? We highlight the limiting assumptions in prior works that make their setup considerably different from the practical setup in which language models are trained. For example, their experimental verification uses ICL objective (training models explicitly for ICL), which differs from the emergent ICL in the wild. Furthermore, the theoretical hand-constructed weights used in these studies have properties that don’t match those of real LLMs. We also look for evidence in real models. We observe that ICL and GD have different sensitivity to the order in which they observe demonstrations. Finally, we probe and compare the ICL vs. GD hypothesis in a natural setting. We conduct comprehensive empirical analyses on language models pre-trained on natural data (LLaMa-7B). Our comparisons of three performance metrics highlight the inconsistent behavior of ICL and GD as a function of various factors such as datasets, models, and the number of demonstrations. We observe that ICL and GD modify the output distribution of language models differently. These results indicate that the equivalence between ICL and GD remains an open hypothesis and calls for further studies.

Details

ICLR Conference 2024 Conference Paper

The Trickle-down Impact of Reward Inconsistency on RLHF

Lingfeng Shen
Sihao Chen
Linfeng Song
Lifeng Jin
Baolin Peng
Haitao Mi
Daniel Khashabi
Dong Yu 0001

Standard practice within Reinforcement Learning from Human Feedback (RLHF) involves optimizing against a Reward Model (RM), which itself is trained to reflect human preferences for desirable generations. A notable subject that is understudied is the (in-)consistency of RMs --- whether they can recognize the semantic changes to different prompts and appropriately adapt their reward assignments --- and their impact on the downstream RLHF model. In this paper, we visit a series of research questions relevant to RM inconsistency: (1) How can we measure the consistency of reward models? (2) How consistent are the existing RMs and how can we improve them? (3) In what ways does reward inconsistency influence the chatbots resulting from the RLHF model training? We propose **Contrast Instruction** -- a benchmarking strategy for the consistency of RM. Each example in **Contrast Instruction** features a pair of lexically similar instructions with different ground truth responses. A consistent RM is expected to rank the corresponding instruction and response higher than other combinations. We observe that current RMs trained with the standard ranking objective fail miserably on \contrast{} compared to average humans. To show that RM consistency can be improved efficiently without using extra training budget, we propose two techniques **ConvexDA** and **RewardFusion**, which enhance reward consistency through extrapolation during the RM training and inference stage, respectively. We show that RLHF models trained with a more consistent RM yield more useful responses, suggesting that reward inconsistency exhibits a trickle-down effect on the downstream RLHF process.

Details

ICLR Conference 2023 Conference Paper

TextShield: Beyond Successfully Detecting Adversarial Sentences in text classification

Lingfeng Shen
Ze Zhang
Haiyun Jiang
Ying Chen

Adversarial attack serves as a major challenge for neural network models in NLP, which precludes the model's deployment in safety-critical applications. A recent line of work, detection-based defense, aims to distinguish adversarial sentences from benign ones. However, {the core limitation of previous detection methods is being incapable of giving correct predictions on adversarial sentences unlike defense methods from other paradigms.} To solve this issue, this paper proposes TextShield: (1) we discover a link between text attack and saliency information, and then we propose a saliency-based detector, which can effectively detect whether an input sentence is adversarial or not. (2) We design a saliency-based corrector, which converts the detected adversary sentences to benign ones. By combining the saliency-based detector and corrector, TextShield extends the detection-only paradigm to a detection-correction paradigm, thus filling the gap in the existing detection-based defense. Comprehensive experiments show that (a) TextShield consistently achieves higher or comparable performance than state-of-the-art defense methods across various attacks on different benchmarks. (b) our saliency-based detector outperforms existing detectors for detecting adversarial sentences.

Details

AAAI Conference 2022 Conference Paper

KATG: Keyword-Bias-Aware Adversarial Text Generation for Text Classification

Lingfeng Shen
Shoushan Li
Ying Chen

Recent work has shown that current text classification models are vulnerable to a small adversarial perturbation on inputs, and adversarial training that re-trains the models with the support of adversarial examples is the most popular way to alleviate the impact of the perturbation. However, current adversarial training methods have two principal problems: a drop in model’s generalization and ineffective defending against other text attacks. In this paper, we propose a Keywordbias-aware Adversarial Text Generation model (KATG) that implicitly generates adversarial sentences using a generatordiscriminator structure. Instead of using a benign sentence to generate an adversarial sentence, the KATG model utilizes extra multiple benign sentences (namely prior sentences) to guide adversarial sentence generation. Furthermore, to cover more perturbations used in existing attacks, a keyword-biasbased sampling is proposed to select sentences containing biased words as prior sentences. Besides, to effectively utilize prior sentences, a generative flow mechanism is proposed to construct a latent semantic space for learning a latent representation of the prior sentences. Experiments demonstrate that adversarial sentences generated by our KATG model can strengthen the generalization and the robustness of text classification models. Benign Sentence Sixthreezero is good, I’ve used it for a long time, only changed because I got tired of the same old bike. (Pos) Prior Sentences S1: Blackberry may work on the systems, but I’m not willing to take that chance on a new expensive phone. (Neg) S2: Iphone4s is in ok previously used condition as stated. But I was disappointed I couldn’t activate the phone upon arrival. (Neg) Adv. Sentence Amazing Iphone4s, used it for so long, only changed because I got tired of the old expensive Blackberry. (Pos) Table 1: Benign sentence, prior sentences and adversarial sentence used in our KATG model. *the corresponding author Copyright © 2022, Association for the Advancement of Artificial Intelligence (www. aaai. org). All rights reserved.

PDF Details