Controllable Reasoning Models Are Private Thinkers
Paper • 2602.24210 • Published
AI & ML interests
Question Answering
Recent Activity
View all activity
haritzpuerto
authored a
paper about 2 months ago
timbmg
authored a
paper 3 months ago
timbmg
submitted a
paper to Daily Papers 3 months ago
timbmg
authored a
paper 7 months ago
haritzpuerto
authored 2
papers 9 months ago
haritzpuerto
posted an update 11 months ago
Post
389
📜 Accepted at ACL 2025! Fine-Tuning on Diverse Reasoning Chains Drives Within-Inference CoT Refinement in LLMs
We propose to fine-tune LLMs to generate diverse chains of thought (DCoT) in a single inference step. This enables within-inference refinement of the cots, no external feedback needed!
🔗 https://arxiv.org/abs/2407.03181
We propose to fine-tune LLMs to generate diverse chains of thought (DCoT) in a single inference step. This enables within-inference refinement of the cots, no external feedback needed!
🔗 https://arxiv.org/abs/2407.03181
timbmg
authored 7
papers about 1 year ago
Best-of-Venom: Attacking RLHF by Injecting Poisoned Preference Data
Paper • 2404.05530 • Published • 1
UKP-SQuARE v2: Explainability and Adversarial Attacks for Trustworthy QA
Paper • 2208.09316 • Published
UKP-SQUARE: An Online Platform for Question Answering Research
Paper • 2203.13693 • Published
UKP-SQuARE v3: A Platform for Multi-Agent QA Research
Paper • 2303.18120 • Published
Incorporating Relevance Feedback for Information-Seeking Retrieval using Few-Shot Document Re-Ranking
Paper • 2210.10695 • Published • 1
Scalable and Domain-General Abstractive Proposition Segmentation
Paper • 2406.19803 • Published • 4
PeerQA: A Scientific Question Answering Dataset from Peer Reviews
Paper • 2502.13668 • Published • 3
haritzpuerto
posted an update about 1 year ago
Post
649
I just got my first ChatGPT review on ARR! 😅 Any advice on how to prove it's AI-generated? Thanks!
- 3 replies
haritzpuerto
posted an update about 1 year ago
Post
1494
I'm excited to announce that my internship paper at Parameter Lab was accepted to Findings of #NAACL2025 🎉
TLDR: Stating an LLM was trained on a sentence might not be possible 😥 , but it is possible for large enough amounts of tokens, such as long documents or multiple documents! 🤯
Scaling Up Membership Inference: When and How Attacks Succeed on Large Language Models (2411.00154)
🔗 https://github.com/parameterlab/mia-scaling
TLDR: Stating an LLM was trained on a sentence might not be possible 😥 , but it is possible for large enough amounts of tokens, such as long documents or multiple documents! 🤯
Scaling Up Membership Inference: When and How Attacks Succeed on Large Language Models (2411.00154)
🔗 https://github.com/parameterlab/mia-scaling
haritzpuerto
authored a
paper over 1 year ago
haritzpuerto
authored a
paper almost 2 years ago
haritzpuerto
updated a
model about 2 years ago
phlippseitz
updated a
model about 2 years ago