Saadi Foundation Books Pdf File

Are LLMs following the correct reasoning paths?

Bangzheng Li, Ben Zhou, Fei Wang, Xingyu Fu, Dan Roth, Muhao Chen,

▶ University of California, Davis ▶ University of Pennsylvania   ▶ University of Southern California
arXiv Code Dataset

We propose a novel probing method and benchmark called EUREQA. EUREQA is an entity-searching task where a model finds a missing entity based on described multi-hop relations with other entities. These deliberately designed multi-hop relations create deceptive semantic associations, and models must stick to the correct reasoning path instead of incorrect shortcuts to find the correct answer. Experiments show that existing LLMs cannot follow correct reasoning paths and resist the attempt of greedy shortcuts. Analyses provide further evidence that LLMs rely on semantic biases to solve the task instead of proper reasoning, questioning the validity and generalizability of current LLMs’ high performances.

saadi foundation books pdf
LLMs make errors when correct surface-level semantic cues-entities are recursively replaced with descriptions, and the errors are likely related to token similarity. GPT-3.5-turbo is used for this example.

saadi foundation books pdf The EUREQA dataset

Download the dataset from [Dataset]

In EUREQA, every question is constructed through an implicit reasoning chain. The chain is constructed by parsing DBPedia. Each layer comprises three components: an entity, a fact about the entity, and a relation between the entity and its counterpart from the next layer. The layers stack up to create chains with different depths of reasoning. We verbalize reasoning chains into natural sentences and anonymize the entity of each layer to create the question. Questions can be solved layer by layer and each layer is guaranteed a unique answer. EUREQA is not a knowledge game: we adopt a knowledge filtering process that ensures that most LLMs have sufficient world knowledge to answer our questions.
EUREQA comprises a total of 2,991 questions of different reasoning depths and difficulties. The entities encompass a broad spectrum of topics, effectively reducing any potential bias arising from specific entity categories. These data are great for analyzing the reasoning processes of LLMs

Image 1
Categories of entities in EUREQA
Image 2
Splits of questions in EUREQA.

saadi foundation books pdf Performance

Here we present the accuracy of ChatGPT, Gemini-Pro and GPT-4 on the hard set of EUREQA across different depths d of reasoning (number of layers in the questions). We evaluate two prompt strategies: direct zero-shot prompt and ICL with two examples. In general, with the entities recursively substituted by the descriptions of reasoning chaining layers, and therefore eliminating surface-level semantic cues, these models generate more incorrect answers. When the reasoning depth increases from one to five on hard questions, there is a notable decline in performance for all models. This finding underscores the significant impact that semantic shortcuts have on the accuracy of responses, and it also indicates that GPT-4 is considerably more capable of identifying and taking advantage of these shortcuts.

depth d=1 d=2 d=3 d=4 d=5
direct icl direct icl direct icl direct icl direct icl
ChatGPT 22.3 53.3 7.0 40.0 5.0 39.2 3.7 39.3 7.2 39.0
Gemini-Pro 45.0 49.3 29.5 23.5 27.3 28.6 25.7 24.3 17.2 21.5
GPT-4 60.3 76.0 50.0 63.7 51.3 61.7 52.7 63.7 46.9 61.9

Saadi Foundation Books Pdf File

A Guide to Saadi Foundation Books: Resources for Persian Language Learners

The Saadi Foundation is one of the most prominent institutions dedicated to the teaching and promotion of the Persian language (Farsi) globally. For students, educators, and Iranologists, the materials produced by this foundation are considered the gold standard for curriculum-based learning. Consequently, the search term "Saadi Foundation books PDF" is highly popular among those looking to master the language without immediate access to physical copies.

Here is a detailed overview of the foundation’s materials and how to access them.

4. The Philological Gap: What the PDF Cannot Replicate

A final, profound layer: The textbook is not the totality of the method. The Saadi Foundation’s pedagogy is built on a teacher-mediated model. The books include listening exercises (requiring CD or download links often stripped from PDFs), group activities, and calligraphy drills. A raw PDF provides the matn (text), but not the savt (voice) or the tamrin-e kelas (class practice). The learner who relies solely on a scanned PDF of the Mīnā workbook without the audio files or the teacher’s guide is like a musician learning from sheet music without ever hearing the melody.

Step 3 — Use proper search queries

Use precise queries in search engines:

The Search for "Saadi Foundation Books PDF"

Due to international sanctions, shipping physical books from Iran to countries like the US, UK, or Canada is expensive and often unreliable. Consequently, many learners turn to digital formats. However, there is a distinct split between legal and illegal PDF distribution.

1. The Institutional Mandate: From Poetic Legacy to Pedagogical Authority

The Saadi Foundation (Bonyād-e Saʿdi), based in Tehran, is not merely a publisher; it is the official, state-sanctioned arbiter of Persian language teaching (Āmuzesh-e Zabān-e Fārsi) for non-Persian speakers. Named after the author of Golestān and Bostān, the foundation carries the weight of classical authenticity. Its textbooks (e.g., the Mīnā series, Fārsi Bebānim, Khoshnavisi series) are designed not by private editors but by committees of linguists from universities like Ferdowsi University of Mashhad and Allameh Tabataba'i University.

The Deep Context: Unlike Duolingo or Rosetta Stone, the Saadi Foundation’s curriculum is ideologically and culturally coded. The books embed not just grammar (syntax, verb conjugation, the colloquial vs. written register split) but also farhang (culture)—including proverbs, Shi'a-influenced social etiquette, and a vocabulary that prioritizes Iranian national holidays and historical narratives. A PDF of a Saadi Foundation textbook is therefore a snapshot of the Islamic Republic’s curated identity, exported globally. saadi foundation books pdf

Availability and Legality

It is important to approach the search for "Saadi Foundation books PDF" with an understanding of copyright and distribution.

Copyright Status: The Saadi Foundation holds the copyright to its materials. Distributing unauthorized PDF copies online is generally a violation of copyright laws. However, because the foundation's goal is the propagation of the Persian language, there are official avenues to access their content.

Where to Find Resources:

  1. Official Website (SaadiFoundation.ir): The foundation’s official website often provides samples, audio files, and sometimes digital access to their content. It is the safest and most ethical source.
  2. Iranology Centers: Many Iranian cultural centers attached to embassies and consulates worldwide are supplied with these books. They may provide PDF materials to registered students.
  3. E-Learning Platforms: The foundation runs an E-Learning institute. By enrolling in their online courses, students often gain legitimate digital access to the textbooks.

Unlocking Persian Literature: A Complete Guide to Saadi Foundation Books (PDF)

For anyone embarking on the journey of learning Persian (Farsi), the name Saadi Foundation is unavoidable. Named after the revered 13th-century Persian poet Shirazi, this organization is the official arm of the Islamic Republic of Iran tasked with teaching Persian to non-native speakers. If you have searched for Saadi Foundation books PDF, you are likely looking for the gold standard in Farsi language textbooks.

But can you find these books legally? Are they effective? And what is the best way to use them for self-study? This article explores the foundation’s curriculum, the availability of digital copies, and the best alternatives for mastering the Persian language.

Acknowledgement

This website is adapted from Nerfies, UniversalNER and LLaVA, licensed under a Creative Commons Attribution-ShareAlike 4.0 International License. We thank the LLaMA team for giving us access to their models.

Usage and License Notices: The data abd code is intended and licensed for research use only. They are also restricted to uses that follow the license agreement of LLaMA, ChatGPT, and the original dataset used in the benchmark. The dataset is CC BY NC 4.0 (allowing only non-commercial use) and models trained using the dataset should not be used outside of research purposes.