AI Papers of the Week
Every paper worth reading in AI, hand-picked one week at a time.
Back to Basics: Revisiting REINFORCE in RLHF
Cohere researchers argue that PPO is overkill for RLHF and that a simpler REINFORCE-style estimator works better in practice.
TestGen-LLM
Meta's TestGen-LLM uses LLMs to improve existing human-written tests - augmenting coverage rather than generating tests from scratch - while rigorously filtering LLM output for quality.
Survey of LLMs
A survey that maps the landscape of the three dominant LLM families - GPT, Llama, and PaLM - and the shared toolbox used to build and augment them.
LLM Agents Can Autonomously Hack Websites
The paper shows GPT-4 agents with tool use and long context can autonomously exploit real websites, including performing blind SQL injection and schema extraction.
LLM-based Multi-Agent Systems Survey
A survey of the fast-growing LLM-based multi-agent systems space, covering both problem-solving applications and "world simulation" research.
Advances in Multimodal LLMs
A comprehensive survey mapping design choices for architecture and training pipeline around multimodal large language models (MLLMs).
Hallucination in LVLMs
A survey specifically scoped to hallucination in Large Vision-Language Models, a phenomenon that differs substantially from text-only LLM hallucination.
WARM (Weighted Averaged Reward Models)
WARM averages multiple fine-tuned reward models in weight space rather than ensembling their predictions, dramatically reducing RLHF inference cost.
Red Teaming Visual Language Models
Introduces the first dedicated red-teaming benchmark for VLMs, covering vulnerabilities unique to multimodal inputs.
Self-Rewarding Language Models
Meta shows that an LLM can act as both actor and judge in its own alignment loop, generating training data without any external reward model.
Sleeper Agents
Anthropic shows that LLMs can be trained to act deceptively under specific triggers and that current safety training techniques fail to remove this hidden behavior.
TrustLLM (Trustworthiness in LLMs)
A 100+ page study that defines a principled framework for trustworthy LLMs and benchmarks 16 mainstream models across it.
Persuasive Adversarial Prompts (PAP)
Turns 40 human-persuasion techniques into a taxonomy of jailbreaks that achieve 92% attack success on frontier models without any optimization.
Mitigating Hallucination in LLMs
A survey cataloging 32 hallucination-mitigation techniques and organizing them into a practical taxonomy.
From Gemini to Q-Star
A 300+-paper survey mapping the state of Generative AI and the research frontiers that followed the Gemini + rumored Q* news cycle.
Exploiting Novel GPT-4 APIs
A red-team study of three newer GPT-4 API surfaces - fine-tuning, function calling, and knowledge retrieval - that reveals each introduces new attack vectors.
Fact Recalling in LLMs
A mechanistic-interpretability study showing that early MLP layers function as a lookup table for factual recall.
Survey of Reasoning with Foundation Models
A comprehensive survey of reasoning with foundation models, covering tasks, methods, benchmarks, and future directions.
Adversarial Attacks on GPT-4
Demonstrates that a trivially simple random-search procedure can jailbreak GPT-4 with high reliability.
FunSearch
DeepMind's FunSearch uses LLMs as a mutation operator in an evolutionary loop to discover genuinely new mathematical knowledge.
Weak-to-Strong Generalization
OpenAI's superalignment team shows that weak supervisors can still elicit capabilities from much stronger models - a first empirical signal for scalable oversight.
LLMs in Medicine
A comprehensive survey (300+ papers) of LLMs applied to medicine, from clinical tasks to biomedical research.
Pearl
Meta's Pearl is a production-ready reinforcement learning agent package designed for real-world deployment constraints.
Llama Guard
Meta's Llama Guard is a compact, instruction-tuned safety classifier built on Llama 2-7B for input/output moderation in conversational AI.