PARAMΔ upcycles dense models to MoE for per-language experts and grafts post-training deltas to enable data-efficient language expansion while preserving original capabilities.
Stanford Center for Research on Foundation Models
6 Pith papers cite this work. Polarity classification is still indexing.
6
Pith papers citing it
citation-role summary
method 1
citation-polarity summary
fields
cs.CL 6roles
method 1polarities
use method 1representative citing papers
TBPO posits a token-level Bradley-Terry model and derives a Bregman-divergence density-ratio matching loss that generalizes DPO while preserving token-level optimality.
NoisyCausal benchmark tests LLMs on causal reasoning with structured noise, and a modular LLM-plus-causal-graph framework outperforms baselines while generalizing to Cladder.
Introduces RevCI benchmark and IMPACT multi-agent framework for evidence-level contradiction detection and graded intensity scoring in peer reviews, distilled into efficient TIDE model.
LIMO achieves 63.3% on AIME24 and 95.6% on MATH500 via supervised fine-tuning on roughly 1% of the data used by prior models, supporting the claim that minimal strategic examples suffice when pre-training has already encoded domain knowledge.
DPOP is a new loss function that prevents DPO from lowering preferred response likelihoods and outperforms standard DPO on diverse datasets, MT-Bench, and enables Smaug-72B to exceed 80% on the Open LLM Leaderboard.
citing papers explorer
-
A Data-Efficient Path to Multilingual LLMs: Language Expansion via Post-training PARAM$\Delta$ Integration into Upcycled MoE
PARAMΔ upcycles dense models to MoE for per-language experts and grafts post-training deltas to enable data-efficient language expansion while preserving original capabilities.
-
TokenRatio: Principled Token-Level Preference Optimization via Ratio Matching
TBPO posits a token-level Bradley-Terry model and derives a Bregman-divergence density-ratio matching loss that generalizes DPO while preserving token-level optimality.
-
NoisyCausal: A Benchmark for Evaluating Causal Reasoning Under Structured Noise
NoisyCausal benchmark tests LLMs on causal reasoning with structured noise, and a modular LLM-plus-causal-graph framework outperforms baselines while generalizing to Cladder.
-
When Reviews Disagree: Fine-Grained Contradiction Analysis in Scientific Peer Reviews
Introduces RevCI benchmark and IMPACT multi-agent framework for evidence-level contradiction detection and graded intensity scoring in peer reviews, distilled into efficient TIDE model.
-
LIMO: Less is More for Reasoning
LIMO achieves 63.3% on AIME24 and 95.6% on MATH500 via supervised fine-tuning on roughly 1% of the data used by prior models, supporting the claim that minimal strategic examples suffice when pre-training has already encoded domain knowledge.
-
Smaug: Fixing Failure Modes of Preference Optimisation with DPO-Positive
DPOP is a new loss function that prevents DPO from lowering preferred response likelihoods and outperforms standard DPO on diverse datasets, MT-Bench, and enables Smaug-72B to exceed 80% on the Open LLM Leaderboard.