Using frontier models to synthesize plausible-but-wrong FIM completions as hard negatives for SFT improves Delulu exact match by +18.8 and edit similarity by +0.22 on Qwen2.5-Coder-7B while also lifting HumanEval-Infilling and SAFIM.
Direct Preference Optimization: Your Language Model is Secretly a Reward Model
4 Pith papers cite this work, alongside 110 external citations. Polarity classification is still indexing.
4
Pith papers citing it
110
external citations · OpenAlex
citation-role summary
background 1
citation-polarity summary
years
2026 4roles
background 1polarities
background 1representative citing papers
SuperMemory-VQA provides 4,853 human-verified QA pairs from 52.9 hours of egocentric AI glasses recordings to benchmark AI systems on realistic long-horizon memory tasks including an unanswerable option.
Counterfactual no-search vs forced-search outcomes yield a model-specific oracle that trains search-routing policies, raising macro-F1 from ~0.71 to ~0.82–0.84 on oracle-eligible examples.
FEST improves RLVR sample efficiency on math and coding benchmarks by combining supervised signals, on-policy signals, and decaying weights on just 128 randomly chosen demonstrations, matching full-dataset baselines.
citing papers explorer
-
Synthetic Hallucinations, Real Gains: Hard Negatives from Frontier Models for FIM Hallucination Mitigation
Using frontier models to synthesize plausible-but-wrong FIM completions as hard negatives for SFT improves Delulu exact match by +18.8 and edit similarity by +0.22 on Qwen2.5-Coder-7B while also lifting HumanEval-Infilling and SAFIM.
-
SuperMemory-VQA: An Egocentric Visual Question-Answering Benchmark for Long-Horizon Memory
SuperMemory-VQA provides 4,853 human-verified QA pairs from 52.9 hours of egocentric AI glasses recordings to benchmark AI systems on realistic long-horizon memory tasks including an unanswerable option.
-
When Should LLMs Search? Counterfactual Supervision for Search Routing
Counterfactual no-search vs forced-search outcomes yield a model-specific oracle that trains search-routing policies, raising macro-F1 from ~0.71 to ~0.82–0.84 on oracle-eligible examples.
-
Boosting Reinforcement Learning with Verifiable Rewards via Randomly Selected Few-Shot Guidance
FEST improves RLVR sample efficiency on math and coding benchmarks by combining supervised signals, on-policy signals, and decaying weights on just 128 randomly chosen demonstrations, matching full-dataset baselines.