Search Results for author:
Found 43 papers, 20 papers with code
Unintended Harms of Value-Aligned LLMs: Psychological and Empirical Insights
Using a dataset with detailed safety categories, we find significant correlations between value alignment and safety risks, supported by psychological hypotheses.
AdAEM: An Adaptively and Automated Extensible Measurement of LLMs' Value Difference
Assessing Large Language Models (LLMs)' underlying value differences enables comprehensive comparison of their misalignment, cultural adaptability, and biases.
Leveraging Implicit Sentiments: Enhancing Reliability and Validity in Psychological Trait Evaluation of LLMs
Through extensive experiments, we demonstrate that: 1) CSI effectively captures nuanced emotional patterns, revealing significant variation in LLMs across languages and contexts; 2) Compared to current approaches, CSI significantly improves reliability, yielding more consistent results; and 3) The correlation between CSI scores and the sentiment of LLM's real-world outputs exceeds 0. 85, demonstrating its strong validity in predicting LLM behavior.
Research on Superalignment Should Advance Now with Parallel Optimization of Competence and Conformity
The recent leap in AI capabilities, driven by big generative models, has sparked the possibility of achieving Artificial General Intelligence (AGI) and further triggered discussions on Artificial Superintelligence (ASI), a system surpassing all humans across all domains.
Benchmarking Retrieval-Augmented Generation in Multi-Modal Contexts
This paper introduces Multi-Modal Retrieval-Augmented Generation (M^2RAG), a benchmark designed to evaluate the effectiveness of Multi-modal Large Language Models (MLLMs) in leveraging knowledge from multi-modal retrieval documents.
PIP-KAG: Mitigating Knowledge Conflicts in Knowledge-Augmented Generation via Parametric Pruning
Knowledge-Augmented Generation (KAG) has shown great promise in updating the internal memory of Large Language Models (LLMs) by integrating external knowledge.
Value Compass Leaderboard: A Platform for Fundamental and Validated Evaluation of LLMs Values
As Large Language Models (LLMs) achieve remarkable breakthroughs, aligning their values with humans has become imperative for their responsible development and customized applications.
The Road to Artificial SuperIntelligence: A Comprehensive Survey of Superalignment
The emergence of large language models (LLMs) has sparked the possibility of about Artificial Superintelligence (ASI), a hypothetical AI system surpassing human intelligence.
Embedding an Ethical Mind: Aligning Text-to-Image Synthesis via Lightweight Value Optimization
To optimize the value encoder, we also develop a framework to automatically construct a text-image preference dataset of 86k (prompt, aligned image, violating image, value principle) samples.
Elephant in the Room: Unveiling the Impact of Reward Model Quality in Alignment
Despite the vital role reward models play in alignment, previous works have consistently overlooked their performance and used off-the-shelf reward models arbitrarily without verification, rendering the reward model ``\emph{an elephant in the room}''.
CLAVE: An Adaptive Framework for Evaluating Values of LLM Generated Responses
The rapid progress in Large Language Models (LLMs) poses potential risks such as generating unethical content.
Raising the Bar: Investigating the Values of Large Language Models via Generative Evolving Testing
Unlike traditional adaptive testing methods that rely on a static test item pool, GETA probes the underlying moral boundaries of LLMs by dynamically generating test items tailored to model capability.
Multi-Evidence based Fact Verification via A Confidential Graph Neural Network
Nevertheless, the noisy nodes usually propagate their semantics via the edges of the reasoning graph, which misleads the semantic representations of other nodes and amplifies the noise signals.
Beyond Human Norms: Unveiling Unique Values of Large Language Models through Interdisciplinary Approaches
Recent advancements in Large Language Models (LLMs) have revolutionized the AI field but also pose potential safety and ethical risks.
On the Essence and Prospect: An Investigation of Alignment Approaches for Big Models
Big models have achieved revolutionary breakthroughs in the field of AI, but they might also pose potential concerns.
Negating Negatives: Alignment with Human Negative Samples via Distributional Dispreference Optimization
Large language models (LLMs) have revolutionized the role of AI, yet pose potential social risks.
ToolNet: Connecting Large Language Models with Massive Tools via Tool Graph
While achieving remarkable progress in a broad range of tasks, large language models (LLMs) remain significantly limited in properly using massive external tools.
LegalDuet: Learning Fine-grained Representations for Legal Judgment Prediction via a Dual-View Contrastive Learning
In this paper, we propose LegalDuet, which continuously pretrains language models to learn a more tailored embedding space for representing legal cases.
ToViLaG: Your Visual-Language Generative Model is Also An Evildoer
Warning: this paper includes model outputs showing offensive content.
CDEval: A Benchmark for Measuring the Cultural Dimensions of Large Language Models
This benchmark serves as a valuable resource for cultural studies in LLMs, paving the way for more culturally aware and sensitive models.
Knowledge Plugins: Enhancing Large Language Models for Domain-Specific Recommendations
In this paper, we propose a general paradigm that augments LLMs with DOmain-specific KnowledgE to enhance their performance on practical applications, namely DOKE.
Value FULCRA: Mapping Large Language Models to the Multidimensional Spectrum of Basic Human Values
The rapid advancement of Large Language Models (LLMs) has attracted much attention to value alignment for their responsible development.
Unpacking the Ethical Value Alignment in Big Models
Big models have greatly advanced AI's ability to understand, generate, and manipulate information and content, enabling numerous applications.
Denevil: Towards Deciphering and Navigating the Ethical Values of Large Language Models via Instruction Learning
We discovered that most models are essentially misaligned, necessitating further ethical value alignment.
From Instructions to Intrinsic Human Values -- A Survey of Alignment Goals for Big Models
Big models, exemplified by Large Language Models (LLMs), are models typically pre-trained on massive data and comprised of enormous parameters, which not only obtain significantly improved performance across diverse tasks but also present emergent capabilities absent in smaller models.
A Survey on Evaluation of Large Language Models
Large language models (LLMs) are gaining increasing popularity in both academia and industry, owing to their unprecedented performance in various applications.
KEST: Kernel Distance Based Efficient Self-Training for Improving Controllable Text Generation
Self-training (ST) has come to fruition in language understanding tasks by producing pseudo labels, which reduces the labeling bottleneck of language model fine-tuning.
Efficient Cross-Lingual Transfer for Chinese Stable Diffusion with Images as Pivots
IAP optimizes only a separate Chinese text encoder with all other parameters fixed to align Chinese semantics space to the English one in CLIP.
Recurrence Boosts Diversity! Revisiting Recurrent Latent Variable in Transformer-Based Variational AutoEncoder for Diverse Text Generation
We demonstrate that TRACE could enhance the entanglement of each segment and preceding latent variables and deduce a non-zero lower bound of the KL term, providing a theoretical guarantee of generation diversity.
Self-explaining deep models with logic rule reasoning
We present SELOR, a framework for integrating self-explaining capabilities into a given deep model to achieve both high prediction performance and human precision.
Unified Detoxifying and Debiasing in Language Generation via Inference-time Adaptive Optimization
Warning: this paper contains model outputs exhibiting offensiveness and biases.
Fuse It More Deeply! A Variational Transformer with Layer-Wise Latent Variable Inference for Text Generation
The past several years have witnessed Variational Auto-Encoder's superiority in various text generation tasks.
CCPM: A Chinese Classical Poetry Matching Dataset
We hope this dataset can further enhance the study on incorporating deep semantics into the understanding and generation system of Chinese classical poetry.
Neural Quality Estimation with Multiple Hypotheses for Grammatical Error Correction
Grammatical Error Correction (GEC) aims to correct writing errors and help language learners improve their writing skills.
Ranked #1 on
Grammatical Error Detection
on FCE
MixPoet: Diverse Poetry Generation via Learning Controllable Mixed Latent Space
Though recent neural models make prominent progress in some criteria of poetry quality, generated poems still suffer from the problem of poor diversity.
Jiuge: A Human-Machine Collaborative Chinese Classical Poetry Generation System
By exposing the options of poetry genres, styles and revision modes, Jiuge, acting as a professional assistant, allows constant and active participation of users in poetic creation.
Stylistic Chinese Poetry Generation via Unsupervised Style Disentanglement
The ability to write diverse poems in different styles under the same poetic imagery is an important characteristic of human poetry writing.
Automatic Poetry Generation with Mutual Reinforcement Learning
Human experts evaluate poetry in terms of some specific criteria, instead of word-level likelihood.
Chinese Poetry Generation with a Working Memory Model
Different from previous methods, our model explicitly maintains topics and informative limited history in a neural memory.
Chinese Poetry Generation with a Salient-Clue Mechanism
As a precious part of the human cultural heritage, Chinese poetry has influenced people for generations.
Generating Chinese Classical Poems with RNN Encoder-Decoder
We take the generation of Chinese classical poem lines as a sequence-to-sequence learning problem, and build a novel system based on the RNN Encoder-Decoder structure to generate quatrains (Jueju in Chinese), with a topic word as input.
